Skip to main content

Full text of "De prosodia hibernica:"

See other formats

P 3 PQUf: 



/ , 

,1 .'■ r 


Digitized by the Internet Archive 
in 2013 


Authore Reverendo Patre 
Fratre Francisco ó Molloy, Ord. Min. Strict. Obs. 

[Rom*:: MDCLXXVII.] 



Friar of the Order of Minors of the Strict Observance, 

at Rome, A.D. 1677, 



Editor of the íaoi Oip'n, the eAccjiA 51011A An -dmAftÁin, 

SeAcc SÁji-óÁncA ^Ae-óitge, 

And Author of For the Tongue of the Gael. 

" Oiait> ah 5AOi-oeAi3 p a me&v rhóft 
A x\-Ái ctiAt nA bf LeAf5 bpion-ot ! " 

Aw r-AcAift pfioinpiAf ó mAotrhtiAi-o (1677). 

Dublin ; 

M. H. GILL & SON, Ltd. 



Printed and Bound in Ireland. 

on W' 






In the year 1677 the Rev. Father Francis ó Molloy (p^omr-iAr 
ó ITlAottiiuAvo), a Franciscan friar, then an exile at Rome, 
published in that city a small work in 12mo. with the title 
Grammatica Latino-Hibernica. This, I believe, was the first 
Irish grammar ever printed, and consisted of twenty-five 
chapters, thirteen of which were on the Accidence of the 
language, and the last twelve on Irish Prosody. It is these 
twelve chapters which are reprinted in the following pages 
with my translation. 

The previous year — 1676 — he had published also at Rome 
a religious work in Irish with a Latin title — Lucerna Fidelium 
— (Lócfv Ann n-A ^C^ erorheAC, or Lamp of the Faithful), a volume 
still occasionally met with and certainly not so rare as the 

These are the only works of our author that appear to 
have come down to our times ; but Anderson, in his Historical 
Sketches of the Native Irish (Third Edition, London, 1846), 
states that Father ó Molloy was also author of an 8vo volume, 
Sacra Theologia, published at Rome in 1667. And in the 
Irish dedication of the Lucerna, addressed to Cardinal Palutius 
Alterius, the author makes distinct reference to earlier works 
which he had written, and for which he had received help 
and patronage from the same exalted ecclesiastic. Some 
of these may still be in existence and may yet be brought 
before the Irish public. The author seems to have been a 
man of some consequence and position at the time in Rome, 
for on the death of the great Father Luke Wadding in 1657, 
Fr. ó Molloy was chosen to succeed him as rector of the Irish 
College of St. Isidore. He was already old (as he tells us 
himself) when he wrote his Irish grammar, but he may have 
lived a few years longer, and his period may be roughly set 
down as from 1616 to 1686. 

The grammar has, I believe, never been translated, though 
often referred to and quoted ; for instance, O'Donovan 

iv Translator's Preface. 

must have thought highly of the Grammatica — though here 
and there he criticizes it freely, sometimes undeservedly — 
for he quotes it in his own Irish grammar no less than eighteen 
times in the accidence and five times in the prosody — 
occasionally to the extent of whole pages, but always in the 
original Latin and never in translation — which, of course, 
looks very learned and very imposing. Though Fr. ó Molloy's 
accidence is quoted so often by 0' Donovan, and though it has 
many points of interest to the Irish grammarian and to the 
student of the Irish language, that part is not near so valuable 
or so systematic as the Prosody ; and the relative value of 
these two subjects in the mind of the author seems to be 
shown in the fact that whilst the whole of the former subject 
takes up but thirteen chapters, the latter alone — as already 
stated — comprises twelve, nearly a half of the book. 

That the Prosody has never been translated before is no 
doubt due to various causes — one being that the work is 
extremely rare — I believe there are hardly three copies known 
in all Ireland. Another, I suppose, is that though a knowledge 
of Latin is common amongst the clergy and the learned 
professions generally, few amongst these classes had any clear 
knowledge or understanding of the technical terms of Irish 
prosody ; and perhaps another reason is — nobody else thought 
of doing it. 

Now this is not a work on the nature of poetry, nor on the 
beauties of poetry, nor on the different kinds of poetry, nor 
does it compare Irish poetry with that of other nations — 
it is simply a technical treatise on Irish versification, as 
practised in the middle ages and down to the seventeenth 

It is not a complete treatise on Irish Prosody ; it does 
not begin early enough, nor with the earliest kinds of Irish 
verse, nor does it come down late enough. Yet the author's 
Cap. XII. De Vulgari Versu ("Of Popular Poetry"), though 
short, is most interesting, and shows that he was familiar 
with the later accented measures, which, indeed, he treats 
as if they were not at all late or recent. In this respect, 
as in a few others, O' Donovan was not quite fair to our 
author, for he not only omits all treatment of such poetry 

Translator's Preface. v 

himself, but leaves the student to infer that Fr. ó Molloy 
had not treated of it. Neither is the original work a well- 
ordered one, or logically arranged ; many things are repeated, 
many other things that one might expect are altogether 

But with all its defects it is a most interesting relic of the 
past — nay, Dr. Hyde, in his Literary History of Ireland (p. 572) , 
does not hesitate to speak of it in unqualified terms as " the 
fullest, most competent, and most interesting account which 
we have of the Irish classical metres as practised in the later 
schools, by one who was fully acquainted both with them 
and their methods." 

I give the text as well as the translation, as scholars, no 
doubt, will be curious to see the original Latin — especially 
as the work itself is so scarce. But as it is no part of my 
duty to correct the author's Latin — even where such might 
seem possible or easy — I give the author's text exactly as 
I find it in the original Roman edition, with the bare correction 
of an odd obvious misprint and the lengthening of one or two 
well-known contractions. Nor am I called upon to criticize 
his style. No doubt, the student fresh from his Cicero or 
his Livy will find a great difference in the Latin of Father 
ó Molloy — the medieval half-colloquial Latinity of the Schools. 
But though in the course of his treatise we may come across 
what looks like an occasional native idiom — such as we find 
in the scholastic Latin of other nations — there is no doubt 
the author had a good grip of the Latin tongue, for not only 
is he able to discuss a technical subject in prose, but in poetry 
also he shows himself competent not merely to write the 
popular medieval riming Latin verses, but also good classical 

And now if it be asked what is the good in these days of 
knowing, or even of troubling about medieval Irish versifi- 
cation, I will reply that — even apart from its value to 
writers who may still love to practise the old metres, and 
two or three of our living poets have successfully shown 
that these metres are not so dead or so obsolete as was thought 
— apart from this consideration, I say, there are many valuable 
reasons for a more general and more accurate knowledge of 

vi Translator's Preface. 

the beautiful and artistic old Irish measures. First, that 
we may the better understand and appreciate our native 
Irish poetry, for in this, as in other arts, it is only those who 
have some technical knowledge that can rightly estimate 
its various points of excellence ; secondly, that we may learn 
something of medieval Irish pronunciation — a matter that 
many know little or nothing of ; thirdly — following from 
the second reason — that we may read Irish poetry with more 
correctness and more expression ; fourthly, that we may be 
able to detect and correct the errors of careless or ignorant 
scribes ; and lastly, that we may more thoroughly and 
exactly understand even the meaning of our old and medieval 
poetry. To many persons — especially the matter-of-fact and 
" scientific " who study poetry not for aesthetic reasons, but 
for the historical or other facts it may convey, this will 
appeal as the most important reason of all. Modern scholars 
have shown that men even so great as O'Curry, O'Donovan, 
Hennessy — nay, great continental savants also, have often 
entirely missed the meaning of lines and passages through 
a disregard or ignorance of Irish Prosody. 

The detached verses and half-verses which the author 
gives to illustrate the Irish metres — removed as they are 
from their context — are in many cases difficult to translate. 
My treatment of these odd verses is to myself — and will be, 
I daresay, to my readers also — the least satisfactory part, of 
my work. Some of them probably are old traditional 
examples often used by poets and grammarians in the schools, 
and going back, perhaps, to medieval times — some of them 
may be as old as the Book of Ballymote (XlVth Century). 
A few are from the writings of known poets as T)ormc.A > o tnó^ 
ó *OÁtAi5, U-a^s T)Att ó li-th^irm, «^ottA-bfuj^oe ó ri-6o$iif^, 
etc. , whilst a good few others are from compositions of Fr. ó 
Molloy himself, and these are generally easy enough. 
Occasionally he gives a Latin version of the Irish examples, 
but for the most part he leaves them untranslated. 
O'Donovan, in his short chapter on Irish prosody, generally 
gives ó Molloy' s examples but leaves them untranslated. 
No doubt, the prosody of such verses is independent of any 
translation, though a reader is better satisfied when he finds 

Translator's Preface. vii 

that the thought equals the expression, and that the 
expression matches the thought. 

The author, at the end of his work, asks his readers' indul- 
gence if — after living many long years away from home and 
books and teachers — he may have erred by excess or defect. 
I, too, his translator, have lived long years away from native 
land — not so many, happily, as the author, and not quite so 
many hundreds of miles away from Ireland, yet I know 
that if I were living in Ireland's capital, I could consult 
friends and books by whose aid the present work would be, 
I will not say more perfect, but less imperfect. Not having 
those advantages, I, too, crave the reader's kind indulgence. 

Little by little and one by one we are gathering up the 
threads of that web that was so violently, so savagely torn 
asunder in the seventeenth century — picking them up and 
trying to piece them together again, ihat so we may realise 
something of the beauty and grandeur of that native Irish 
civilisation that was then thirty centuries old. It was the 
most famous century in all our late history, full of brave 
deeds and brilliant men — men of thought, men of action — 
a century that saw three great struggles for Irish liberty, 
and had enjoyed brilliant if brief periods of liberty. We do 
not know as yet a hundredth part of the life and characters 
and activities of that strenuous heroic time. 

UonUf ó "PLArm&dHe. 


Cap. I, De Prosodia et Carminum generibus 

„ II. De Quartis et Syllabis . . 

„ III. De Concordia . . 

,, IV. De Correspondentia et Extremis . . 

,, V. De Metro Hibernis TDeibi-oe 

„ VI. De Metri genere Hibernis Seu-otiA 

,, VII. De Metro magno et parvo necnon CArbAifi 

,, VIII. De Carmine vulgd ÓstÁcAf 

,, IX. De Carmine vulgd "OfioijjtieAc 

„ X. De Carmine Hibernis bjiuilinjeAcc 

,, XI. De quibusdam prcecognitionibus . . 

,, XII. De Vulgari Yersu, et Compositione 



Cum An Ai<\\\ pÁ-ofiAic O^iAt 
"Don pUuncéA-oAc lAfitA pne gAtl 
Ad Rev. Patrem Patricium Tyrellum .,. 
An t>Áp : *OonncAX) xr\óp ó "OÁtAij *oo cum 
Softvut) ó "óítjieAbAc ttuAmA 50 CtÁfi Coitin 
Sofiviit) 50 h-AOf 05 "( eACA Oitém tiA tlAom 



Chapter I. Of Prosody and the kinds of Metre 
II. Of Quartans and Syllables 

III. Of Concord or Alliteration 

IV. Of Assonance and of Terminations 
V. Of the Metre called by the Irish TjeiDi-oe 

VI. Of the Metre called by the Irish Seu-otiA 
VII. Of the Greater and Lesser ilAnninse^ct: and of Car-bAi 
VIII. Of the Verse commonly called ójtÁc^f 
IX. Of the Verse commonly called *Ofioi5neAc 
X. Of the Verse known to the Irish as bfiuitinjeAcc 
XI. Of certain things to be known beforehand . . 
XII. Of Popular Poetry, and of Composition 



n 53 


De Prosodid Hibernicd. 


Authore Rev. P. Fratre Francisco ó Molloy, 
Ord. Min. Strict Obs. 

[Roma?: MDCLXXVII]. 
CAP. I. [xiv]. 

1. Hucusque grammaticum utcumque egimus ; literas, 
syllabas, orthographiam, pronunciationem, declinationes, 
conjugationesque insinuavimus, necnon et orationis partes. 
Animus erat Rhetoricam praemittere Poési. Poeta enim 
laureá dignus ex prosá seu soluta oratione ligatam faciens, 
suos eleganter concinnat versus, et vix aliter : verimi non 
dabatur tempus, alio quippe et alio distrahebar, corpore 
inflrmus properante ad occasum. Solam igitur dabo Proso- 
diam veterum regulis et observationibus refertam. 

2. Prosodiam hie usurpo pro methodo quádam seu arte 
construendi carmina. Carmen autem oratio est strictiori 
pedum seu syllabarum lege rité coércita ; ornatissimum 
dicendi genus. Carmen apud Hibernos est triplex, scilicet 
metrum vulgo x>Án 'oípe&ó, et DiMntingeAóc, et ógtÁóAf, de 
quibus infra. Maximé autem de metro, omnium quae 
unquam vidi vel audivi, ausim dicere quae sub sole reperiuntur, 
dimcillimo, quo fnimirum bene semel cognito, nulla in reliquis 
cognoscendis supererit dimcultas. 

De Prosodid Hibernicá. 


By the Rev. Father Francis ó Molloy, 
Friar of the Order of Minors of the Strict Observance. 

[Rome: 1677]. 
CHAPTER I. [xiv]. 

1. Thus far we have in some fashion written as the gram- 
marian : letters, syllables, spelling, pronunciation, declensions 
and conjugations we have touched upon, as well as the parts 
of speech. It was my intention to prefix a treatise on Rhetoric 
to the Prosody. For a poet deserving of the laurel, making 
measured speech out of unmeasured, composes his verses 
with much art, and with difficulty can he do without such 
art.* But time did not permit; by one thing or another 
I was distracted, I was growing infirm with a body hastening 
to decay. The Prosody alone, therefore, I will give here, 
accompanied by the rules and observances of the ancients. 

2. By Prosody I here mean a certain method or art of 
constructing verses. But verse is speech duly restrained 
by a more or less strict law of feet or syllables — the most 
ornate mode of expression. Verse or metre, with the Irish, 
is as commonly given threefold — namely -oÁn t)íj\eAC (direct 
metre) uruiilinse-aoc, and ó^tÁCAf, all of which will be treated 
of below. But especially I must treat of that metre — -oÁn 
■oíjteAó — the most difficult of all I have ever seen or heard, 
and I durst say of all that are now found under the sun ; 
which being once well understood, no difficulty will be found 
in understanding the rest. 

* The author implies that even a poet would be the better for a 
knowledge of the art of rhetoric. 

4 De Prosodiá Hibemicá. 

3. Carmen hoc ut evadat metrum, Hibernis "oÁn -oipeAt 
vel nann "oineAC, septem necessario expostulat : certum 
scilicet syllabarum numerum, quartorum numerum, con- 
cordiam, correspondentiam, extrema seú terminos, unionem 
et caput ; quae vulgó dicuntur nuirhif\ CeAt^orhAn, cmceAóc 
fioVUvo in -&AÓ ce^tnoniAin, UAim, comA^Att, finn -A^uf 
Aijvojunn, uAitne A^uf ceAnn. His octavum addi possit, 
non quod semper sit necessarium, sed quod frequenter admit- 
tatur, vulgó ujvunn, de quo suo dicemus loco. 

4. Genera metrorum prsecipua (ut omittam minus princi- 
palia, vulgó coft\-AifT)e, seu -po-Aifoe, quae varia sunt), et 
principaliora, ac nunc magis in usu apud Hibernos, sunt 
quinque, vocanturque -oeiDroe, yeAX>r\A, -p-Annvnje-Acc tie^s, 
fVAnnvnjeAcc rhójA et CAfDAifw. 

5. Adverte autem ex necessariis septem supra numeratis, 
quatuor priora scilicet numerum quartorum, numerum 
syllabarum, concordiam et correspondentiam — hoc est ntnrhin 
CeAtporhAVi, muring fioLUvo m ^aó ceAtpomAm, uA\vn if 
cotri^^A-o — requiri indispensabiliter ad quodcumque metrum 
cuj usque fuerit generis. Verum non sic de tribus 
ultimis necessariis ibidem numeratis ; quia requiruntur 
necessario non ad omne genus, sed ad qusedam genera. Sic 
majus et minus extremum requiruntur ad genus vulgó -oeibroe 
praecise et indispensabiliter ; unto veró non nisi ad nAnmu- 
$e.ACC tfión, et cdftiAi|\n ; ce^nn autem ad -pAnntnje^cc tte-dj; 
et f eAvriA solummodó. 

De Prosodid Hibemicd. 5 

3. In order to have this kind of verse — called by the Irish 
•oÁn -oípeAó, direct metre, or pAnn "oipeAC, direct verse — 
seven conditions or requisites are necessary : (1) the due 
number of quartans ; (2) a fixed number of syllables (in 
each quartan) ; (3) concord or alliteration ; (4) correspondence, 
i.e., assonance or rime ; (5) point or termination ; (6) union, 
and (7) ending ; which with us are called respectively mutiny 
ceAtporhAn, anceACc fioLUvó, tiAnn, cotiiAjvoAt), furrn (4511 p 
Ái-fvofurm), uAitne, Ajjup ceArm. To these may be added 
an eighth — not that it is always necessary, but that it is 
frequently admitted — commonly known as ujaAnn (shaft or 
handle) of which we will speak in its own place. 

4. The chief kinds of metre (to omit the less important, 
such as those commonly called cojXjA-Aipoe* or po-Aifoe, of 
which there are many varieties) — the most important and 
those now most in use with the Irish are five, and are called 
'oeiDi'óe, feA^nA, pArmtnjeACc oeAg, pAtirmijeACu tfióp and 

5. Notice, however, that of the seven necessary requisites 
above enumerated, the four first are indispensably required 
in every metre of whatever kind it may be — namely, the 
due number of quartans, the right number of syllables, 
alliteration and assonance — in Irish the nuirhip ceAtponiAn, 
rmitíiifA fiottAt) m jac ceAtporhAin, uAirn and cotfiAjvoAt). 
But it is not so of the three last requisites there mentioned, 
for they are necessary not for every metre, but only for some. 
Thus the greater and less point (or termination) i.e., jurm, 
are certainly and indispensably required in the metre called 
'oeitM'óe, uAitne or union however only for nAntunjeACc riióp 
and cA-poAifm, whilst ceAtin or ending is needed only in f\Armui- 
£eAóc oeAg and feA-oriA. 

* Cojift-Aipoe=odd o r occasional metre, 1?o-Air-oe=:sub-metre or 
secondary metre. These words show that Aipoe (or Aipce) was also 
a well-known term for a metre or metrical form. It came also to mean 
a poem in general, a composition in verse — example of the general 
from the particular. Compare the English terms rime, metre, numbers, 
verse, etc. 

6 De Prosodiá Hibemicd. 

CAP. II. [xv.]. 
De quartis et syllabis. 

1. Metrum complete sumptum vulgó j^nn seu t^nn lomLÁn, 
construitur ex duobus semimetris. Quartum vocatur in metro 
sermo constans pluribus dictionibus co-adunatis conflantibus 
determinatum syllabarum numerum. Appellatur quartum 
quia est quarta pars metri completi, tametsi sit unica pars 
ex duabus quibus conflatur semimetrum. Poéma vulgó 
•oÁn multis constat metris, et tot quot voluerit author ; 
aliis T)tiAin. Quando ex duobus semimetris vulgó leAtfArm, 
integratur metrum, primum semimetrum Hibernis nuncu- 
patur f eoUvo, secundum veró corh^T). 

2. Quodlibet metrorum apud Hibernos in poémate debet 
secundum se perfectum claudere sensum et orationem sine 
dependentiá ab altero. Imó primum semimetrum indepen- 
denter a secundo perfectum generat sensum : nihilominus 
quando ex duobus ut supra conflatur metrum, sese invicem 
respiciunt et eorum sensus mutuó referuntur ad idem pro- 
positum. In componendis autem metris bene incipitur a 
secundo semimetro vulgó cotfiAt) ; ut cum pleno sensu et 
nervo gravius et gratiosius fieri possit primum semimetrum, 
eique quadrare; longé quippe dimcilius fit corhá-o quam 
feolA-o ut patebit infra. 

De Prosodiá Hibemicd. 7 


Of Quartans and Syllables. 

1. A metrical verse taken as a whole — commonly called 
a jiArm, or f\Ann lotnlÁn — is made up of two semi-metres or 
half- verses. In verse that much of a metre is called a quartan 
which, consisting of several words united, contains a definite 
number of syllables. It is called a quartan because it is a 
quarter or fourth part of the whole verse (or stanza), whilst 
it is one part of the two forming a semi-metre. A poem, 
which we commonly call t>ati, consists of many verses 
(or stanzas), as many as the author may wish ; it is by some 
called a "ouaiti.* A semi-metre is called a teAt^Arm or half- 
verse, and when of two semi-metres a full verse is made, 
the first is called by the Irish feotA-o or the " leading," the 
second coííia-o or the " closing." 

2. Whatever the number of verses in a poem, they should 
in themselves make and include a complete sense without 
depending on any other poem or division of a poem. Nay, 
even the first semi-metre of a verse should make a complete 
sense, independently of the second ; nevertheless, when, as 
above stated, a verse is made of two parts, they may be related 
to each other, and their sense be mutually referred to the one 
proposition. In composing verses, however, it is well some- 
times to begin with the second semi-metre (!) i.e., with the 
corhA-o ; so that the first semi-metre may be made with 
full sense and strength, and match the second one the more 
worthily and the more gracefully ; for the coxx\ax> or second 
half is far more difficult than the first, as shall appear below. 

* But -otiAri (or -ouAin) originally was a special metre — indeed it was 
but another name for jiAnn-uijeAcc beAt;, whilst jiAnnvnjeAcc itióp 
was called the fOfi-T)t)An or greater -ouAn. In course of time the word 
came to be used for a poem in general, especially a poem of distinction, 
thus T>uAn móp-ÓA was used for an epic poem. Hence -ouAnAifte, a book 
or collection of poems. 

8 De Prosodiá Hibernicd. 

3. Numerus syllabarum vulgó córh^f vel cinceACx f ioIU* 
est quodlibet metri quartum constare septem syllabis non 
pluribus — sine eo quod aliqua elidatur — neque paucioribus 

lonró-A fgeuL mAit a\\ TVIuijAe — 

Excipe tamen metrum generis nuncupati \>éAX>r\& in cujus 
primi et secundi semi-metri quarto requiruntur octo syllabae, 

TTlAifu; feuóAf Af 1nif-ceittiotin ! 

Dixi sine eo quod aliqua elidatur seu mergatur, ut dictum 
est de mersione vulgó t>Át<yó ; quoties enim ex duabus syllabis 
pronunciari debet et fieri unica, ut interveniat hsec elisio, 
requiritur quod aliqua vocula finiat in vocalem vulgó cotiiAj\- 
T)A1$. Debet insuper intervenire adverbium vulgó lApmtoeufaA, 
incipiens cum vocali, et sequens immediate ad prsefatam 
voculam, ut finalis voculse vocalis et initiativa adverbii 
invicem elidant, ita ut unica efferatur syllaba ordinarié 
brevis ; ut, 

AriAro j\e fojA a ílí tíiáija ! 

etenim a in fio&d per a subsequens adverbiale eliditur. 

De Prosodiá Hibernicá. 9 

3. The number of syllables, called cottier* (metre or 
measure) or cmce^cc f ioUatd (definite complement of syllables) 
is seven in each quartan (ceAt^Am or line) not more, except 
when one is elided, nor less, as : 

1otnt)A fjeut iridic Ap tnuif e : 
[i.e., Good the tales told of Mary !] 

Except,f however, in the kind of metre called yéAX)t\A, in the 
first line of each semi-metre of which eight syllables are 
required, as, 

1T1ai|\5 -peuc^-p Aj\ 1nif Ceicttonn ! 
[Sad we gaze on Inniskillen !] 

I have said except when a syllable is elided or merged by 
the figure called bÁtAt> (merging or sinking), for whenever 
two syllables should be pronounced as one, in order that 
this elision may arise, it is necessary that one word should 
end in a vowel [called the corhAjvo-Aij; ?]. There should also 
occur an iAj\rnbeufa.A or adverb beginning with a vowel and 
immediately following the aforesaid word, so that the final 
vowel of the one and the initial of the other meet and form 
one single syllable, usually short ; as, 

Ar\Ató j\e tvoj-A a RÍ triAi|\J — 

[Wait for the chosen one, O great King !] 

— where the a of ^oj^ is elided by the subsequent particle a. 

* This word róúiAf shows that the original and fundamental dis- 
tinction between metres in Irish was the measure of the line, that is, 
the number of syllables in it. Indeed the word (formed from -oo-fo- 
mefr) is a compound of mer-r- now tneAr-, and contains the same root 
as {Ler-pov and the Latin met-or, met-ior, mens-ura, whence "measure." 
Other circumstances besides the number of syllables, no doubt, in later 
times gave rise to varieties of the same metre. 

f The author overlooks the five-syllabled metre called -oeAcnA-ó beAj; 
and the six-syllabled jiiorinAjfo in which the -peiti^e xionjufA was 
written. There were also four-syllabled and even three-syllabled 
metres — whilst on the other hand, -oftAijjn eAc admitted any number 
of syllables from nine to thirteen. But in -oeibi-oe, cAfbAi^n and 
the two species of fiAnrmi jeAcr, the seven-syllabled line was the rule. 

I The word I have read mÁifi is doubtful. It may have been meant 
for tflÁiftfi (a fií rhÁinfi, O king of Marr I). It seems misprinted in the 
Roman edition, which has " nmhAi|V|i." 

10 De Pro so did Hiberntcd. 

4. Item: 

An ca£a S^AjAjA Ay Í Ar treA|Aj\ : 

— ubi in scansione ex a brevi in fojA et a adverbiali sequenti 
fit una syllaba ; item ex í et a in Af unica fit syllaba. Verum 
quando finalis vocalis voculae, ut dixi, corresponsalis est 
brevis, initiativa vero subsequentis adverbii est longa, tunc 
potest ibi indifferenter fieri vel non fieri mersio, juxta exigen- 
tiam quarti, ut videre est in — 

puAf Af 005A ó t)]MAn DUTOe : 

— ubi ad servandas septem dumtaxat syllabas illaesas integré 
ex a brevi in bojA et ó subsequenti nulla fit elisio, 
omittitur autem elisio ne desit syllaba ; alias tamen 
melius elideretur, tametsi excusetur elisio aliter facienda, ut 
dixi, certá quádam licentiá. Dixi melius, quia altera voca- 
lium est brevis, altera longa ; unde cum utraque quandoque 
est brevis, nulla fit elisio, ut patet respective ad a in pa et 
1 in i4f\5nó ut hie : 

AcAim pA lAfgnó ón éAóc ! 

De Prosodid Hibemicd. 11 

4. So also in — 

An za%a seAffv a\> í Ay peÁ^ : 
[The quicker the choice the better !] 

as in scansion the short a of fojA and the adverbial a 
following make one syllable, so also from 1 and the a in a? 
one syllable is formed. But when the final vowel of a word, 
as I have said, is short, but the initial of the next is long, 
there may or may not be an elision in that case, according 
to the exigency of the verse, as may be seen in — 

pu^^f bojA ó t)j\i<Ati bui'óe^ — 
[A bow I had from swarthy Brian] — 

where to keep at least seven syllables no elision is made from 
the meeting of the a in bojA and the preposition o, and the 
elision is omitted lest there should be a syllable wanting. 
Otherwise it would be better to have elision, although it may 
be excused when it could be done only, as I have said, 
by a certain licence. I have said it would be better, because 
in this case one of the vowels is short, and the other long ; 
whence whenever both of the vowels are short no elision 
occurs* as appears when we look at the a in jm and the 1 of 
i^pio in the following — 

AcÁitn -p-A i^jxgnó ón éAóc ! 
[I am in anguish from the death] — 

* Here the author seems to contradict himself, for in a previous 
example ,dnAix> jte \\0£& a ftí íiiÁi|i — two short vowels meeting cause 
an elision. But though in such case bÁiA-6 or elision is usual it is 
not of obligation. 

12 De Prosodiá Hibernicá. 

CAP. III. [xvi.]. 

De Concordia. 

1. Concordia vulgó uAim, duas expostulat voculas, quarum 
neutra sit adverbium, vulgó lAfrnibeutAUi, in omni quarto 
hujusmodi metri — vulgó patin trifieAÓ — quae voculae debeant 
indispensabiliter incipere, vel ab aliqua vocali, hinc inde 
ejusdem vel diverse speciei, vel ab eádem omnino consoná, 

lotnt)^ f^etit m^it Aft tTluifie, 

¥& tnoLcAf\ a tníoj\í)tiite — 

T)o §eiu a\^ aw 015 n-icoum, 

Sjét av cóij\ *oo c|\eToiorhtiin. 

In primo enim quarto concordant niAit et tTluipe, utpoté 
incipientia ab m ; similiter moVCAp et tnio^ouite incipientia 
ab m, concordant in secundo quarto ; in tertio autem 015 
et 10-otnn propter o et 1 vocales initiativas, nee obest n in 
n-io-otnn, quia non est litera propria seu possessiva istius 
voculse, sed accidentaria et adventitia, ad quam non debet 
attendi. sed ad sequentem hujusmodi adventitiam, qualis 
ibidem est 1. Similiter concordant in quarto seu ultimo 
quarto dicti metri coija et c^eroiorhtun, propter initiativas 
nempé c utrinque repertam. 

De Pro so did Hibernicd. 13 


Of Concord or Alliteration. 

1. Concord or alliteration, called by us iiAim, requires that 
two words of which neither is an iA|\tnbeti|\tA or adverb 
should be found, in each line of this metre, beginning indis- 
pensably either with a vowel — of the same or of different 
species — or with the same consonant ; as, 

lonró-A rseut mA\t ajv trim^e,* 
P-Á tnoLc<dj\ a miof\Diiite — 
T)o jeio A\y aw 015 11-10*011111, 
Sgét Af cchf T)o ófeiTnorhuiti. 

Here in the first quartan (or line) triAit and mui|\e agree in 
their initials, that is, they accord or alliterate — beginning 
with the same consonant m ; similarly motcA^ and mio^tnte 
beginning with the same letter m agree in the second quartan ; 
in the third also 015 and 10*011111 agree in having initial vowels, 
nor does the n in the beginning of ti-io*óum prevent this, 
because it is not a proper or radical letter of this word, but 
only accidental and adventitious, f which need not be con- 
sidered, but rather the succeeding radical vowel 1. So also 
in the fourth or last quartan of the above verse, cóifv and 
cneiT)iorhtiiti agree or alliterate as they begin with the same 

* Literally : 

Many's the good story of Mary, 
In which her miracles are praised — 
The reader finds of the pure Virgin — 
A story 'tis right to believe. 

t This n — usually called an eclipsing n — is really in this case a relic 
of the accusative ending (prehist. Irish — ax\) of the noun 015 (for 
013-en, Ó3-AT1) governed by the prep. 4ft. In modern Irish this n 
would not be used. 

Though a succession of two (or more) words beginning with vowels 
is referred to as a species of alliteration, and may be called a general 
vocalic concord, this, of course, does not mean that the vowels are 
considered of the same value or sound, for in accented riming syllables 
the vowels never interchange. It is to be noticed that, in initial 
concord, the vowels are to be different — I believe, therefore, that with 
regard to initial vowels what was sought after was melody or variety. 

14 De Prosodiá Hibernicá. 

2. Haec autem concordia est duplex, propria scilicet et 
similitudinaria, vulgó piojvuAim et uAun jnúife. Propria 
dicitur ilia qua duae ultimae voculse alicujus quarti concordant 
modo jam dicto. Similitudinaria autem, qua duae voculae 
alicujus quarti concordant quidem, etsi non sint ultimae 
ipsius dictiones. Exemplum habes concordiae propriae in 
singulis quartorum allati jam metri. Exemplum autem 
similitudinariae accipe : 

T)'fMOf co^Ait) corhAitceAf potóÁin : 

— similitudinariae enim modo jam explicato concordant 
cos-dift et corhAitce-dfi incipientes, esto non sint ultimae 
dictiones quarti, imó ut sic concordent hac concordantia 
necessum est non sint ultimae. 

3. Adverte autem quod adverbium nunquam facit con- 
cordiam nee ipsum impedit. Similiter neque corresponden- 
tiam facit — de qua infra — neque impedit, neque majus extre- 
mum neque minus, neque unionem, de quibus infra, unquam 

De Prosodiá Hibefnicd. 15 

2. Now this concord or alliteration is twofold, namely 
proper (or true) and improper or apparent, or as we call 
them -piofv-u,dim and u.dim §núif e. That is called piofv-uAim 
or true alliteration in which the two last (chief) words of 
a quartan agree in the aforesaid manner. It is called uAim 
gnfiife — agreement to the eye — when the words of a quartan 
alliterate, but are not the two last words. Of the true or 
proper alliteration you have an example in each line of the 
stanza already given. Of the improper or apparent allitera- 
tion take this : 

TD'-pio^ Ó05AIT) corhAitce^f riotóáin : 
[By the man of war is fostered peace.] 

Here co^avd and corhAitceAjA agree in their initials in the 
manner already explained, but as they are not the two last 
words of the line, the concordance or alliteration is the 
apparent* or improper species. 

3. Notice, however, that an adverb] never makes or hinders 
an alliteration. So also it neither makes an assonance (or 
rime) — of which more below — nor hinders one, nor does it 
ever constitute a j\inn or point, the greater or the less, nor 
a union (internal rime), of which matters I shall treat hereafter. 

* Similitudinaria (apparent or seeming) is here not a very exact 
term, for cojai-ó and coTriAitceAji make as true and real an alliteration 
as any other two words beginning with the same consonant — the 
aspiration of c (in 60541-6 ) not affecting the matter, as is explained 
later. The only difference is, they are not in the strong or emphatic 
place — for as the second half of a verse was usually the stronger and 
more emphatic half, so also the second half of a line was treated as 
the stronger half. Strong and weak would better characterize the 
two positions. 

f Adverb t Under the head adv erbium the author includes a great 
many different kinds of particles and undeclined words, besides what 
are usually known as adverbs ; for instance, the article, possessive 
pronouns, the relative pronoun, the simple prepositions, conjunctions, 
and even the verbs if, b<\, bu-6, buf — in fact, all short and unaccented 

16 De Prosodiá Hibernicd. 

4. Concordia quae dicitur propria, servire potest loco 
similitudinarise, et vice versa in omni quarto, prater tertium 
et ultimum secundi semimetri, qualia semper et indispensa- 
biliter propriam requirunt concordiam, sine qua nunquam 
sufficit similitudinaria, quae multoties sumcit in primo semi- 
metro. Non requiritur prseterea adverbium tametsi toleretur 
jacere ante voculas similitudinariae Concordes in quarto, ut 
videre est in hoc : 

T)o coriiAf cotfixvo tiÁ^ rhAit : 
ubi adverbium «oo toleratur, nee similitudinariam impedit 
concordiam inter subsequentes voculas. Poetae tamen 
despicienteshujusmodi similitudinariam adverbio postpositam, 
voeant asperam, vulgó uAitn $aj\d, vel ^núif %Af&. Tu autem 
hanc evites velut ingratam, sed ut evitare queas, sequentes 
tibi prsescribuntur regular : 

5. Primó, ut ultima quarti vocula sit nominativus verbo, ut 

T)'po|\ cogAit) corhAitce^fv fíotóÁm : 

Secundó, vel ut verbum iiniat subsequenter ad nominativum, 

An coga'O ceA|\c DtiA-ó-ái^e^f ! 
Tertio, vel ut adjectivum iiniat post substantivum, ut, 

5-ADAm cu$;.dm cormjv<vó m^ic ! 
Quarto, vel ut substantivum ad adjectivum iiniat, ut — 
Hi ^edf fWft -Af triAit cormjv<yo. 

De Prosodid Hibernicd. 17 

4. The proper alliteration can, of course, serve in the place 
of the improper, and the latter for the former, in every line 
except the third and fourth of the verse, which always indis- 
pensably require the true alliteration — and here the improper 
(or weak) never suffices, although it is often sufficient in the 
first semi-metre. Nor is there any adverb required, although 
one may be allowed to stand before the improper alliteration 
in the f eoLd-ó or first semi-metre ; as may be seen here : 

T)o cofnAf corhxvo r\Áp ttiAit — 
[I have defended a gift that was not good]. 

— where the adverbial particle "oo is tolerated, but does not 
hinder a minor alliteration between the two subsequent 
words. Poets, however, not approving of this kind coming 
after an adverb, call it tj^im g^u or the rough alliteration, 
or sometimes snúif $Aftf>, the rough aspect. You, however, 
must avoid this unpleasant kind, but in order that you may 
be able to do so, the following rules must be prescribed : 

5. First, the last word of a quartan may be the nominative 
to the verb, as in — 

X)'pop 003^1-0 corhAitce^f fíotcÁitv (Sé-d'ónA.) 
[By the man of war peace is fostered.] 

Secondly, a line may finish with a verb following a nominative, 

An cosaí) ce&^z XyuA-oMgeAif ! 

[Tis the just war that triumphs !]. 

Thirdly, it may finish with an adjective following a substantive, 

^ADAm cu^m cormjvdt) m^it ! 

[Let me accept a good bargain.] 

Fourthly, the line may end with a noun dependent on an 
adjective, as in, 

tli fe^jt mifi -Af mxMt cormtv<vo — 
[I'm not a man good at a bargain !]. 

18 De Prosodiá Hibernicá. 

Quinto, vel ut substantivum quod in genitivo regitur ab 
alio substantivo fmiat, ut, 

ílí COtfvOtn COJAlt) X)ATiX)A 

Sextó, vel ut verbum activum finiat post accusativum, ut, 

Af bpeAj'óA An tte^n foin toicnn — 
Septimó, vel ut substantivum in accusativo íiniat post verbum, 

ílí beo'óA íaoó toiceAf mtiAoi ■ 

6. Neque eclipsis nec aspirativa n, neque nudatio seú 
obtenebratio — vulgó uijvóui&a-ó de quá suprá, neque f éirhmjAt) 
neque tom^o impedit hanc concordiam, nisi in sequentibus ; 
quando scilicet litera n sequitur immediate ad p et efncit 
ut efferatur instar p, et sic litera p erit initialis alterius voculae 
concordantis cum dictione incipiente pti, ut videre est in hoc 
quarto : 

A*oiriAim -ÓU1U mo prieACAro pém ! 

Quoties autem n sequitur ad consonam p voculae initiativam 
toties ilia consona -p non erit quae facit concordiam, sed prima 
litera subsequens ad p sic aspiratam seu mortificatam per n, 

Za^am^ le-Atn a pUMt éijwe ! 

ubi t in leAm et t in ptAit faciunt concordiam similitudi- 
nariam, nulla habita ratione istius pn. Item 

Ua5aij\ teAtn a ptAit Upe — 

— ubi t in fiA\t et t in tipe propriam faciunt concordiam. 

De Prosodiá Hibernicá. 19 

Or, fifthly, it may finish with a genitive dependent on another 
substantive, as, 

Hi cotfiom C05-A1-Ó t)AuX)A ! 
[Unequal the fighting in Banva !] 

Or, sixthly, an active verb after an accusative may end a 
line, as in, 

Af bfeAj-óA An be^n fom toicim : 
[Fair is that woman I injure.] 

Or, seventhly, a substantive in the accusative to an active 
verb may end the line, as in, 

TH beo"óA Laoc toiceAf tntiAoi ! 
[Unmanly the soldier that injures a woman !] 

6. Neither eclipsis nor aspiration, nor the nudation* of a 
consonant (t, n, r\) that is, neither uijvóiugA'óf of which we 
have spoken above, nor peitfiiu&vo nor lomA-o — hinders an 
alliteration, except in the two following cases : 

(1) When the letter ii immediately follows a p and causes 
it to be pronounced like p, then also -p (or pn) must be the 
initial of the alliterating word, as may be seen here : 

xVotiiAim t>uic mo peACAit) -pew. 
[I confess my sins to thee !] 

(2) Whenever b follows the consonant p in the beginning 
of a word, that is, when an initial p is aspirated, then it is 
not this p which is to make the alliteration, but the next 
consonant (t or r\) — if a consonant follows — as in 

Ua5-ai|\ leAtn a ptAit éirme ! 
[Prove to me, lord of Erne !] 

— where the t of teAtn and the t of -ptait make an apparent 
concord or alliteration. Also 

UAgAifv teAtn a ftAit Upe — 
[Prove to me, O lord of Liffey !] 

— where the t in £UMt and the t in bpe make the proper 

♦The author does not define nudatio or tomA-ó in his Grammatica, 
but I think he means the lightening of t, n, ji at the beginning of words 
when under the influence of aspirating particles, mo, -oo, etc. In 
tÁTh the 1 is rjiom or rearm, in mo "LÁm the I is éA-ojiom or torn (light 
or bare). 

t UifróiujA-ó (eclipsis) was sometimes considered a derivative of 
■oíobA-ó or -oíojA-ó cutting off, destruction, O. Ir. "oibxvo, but more 
usually it is written uft-x>vib,<vó, a compound of -oubA-o darkening, from 
■oub, black, dark. 

20 De Prosodiá Hibernicá. 

7. Litera f initialis numquam concordat nisi cum alia f 
et tali qualis ipsa, seú qualiter ipsa afficitur, adeoque p non 
concordat cum fb, neque cum re, neque cum fo, neque 
cum rs, sed -p simplex requirit f simplex. Similiter ft) 
postulat fb, et sic de aliis ut interveniat concordia. Pari- 
formiter non concordat cum cf post particulam An : sed 
cf ut concordet requirit aliud cf, quale postulat articulus 
An, ut alibi diximus, exemplo de An Cfúit, An cftije, etc. 

8. Vidimus alibi consonantes molles numero tres, videlicet 
c, p, c ; item duras nempé b, r>, 5 ; item asperas scilicet 
en, pn (p), en ; item quinque fortes ut It, nn, |\n, m, n^ ; 
item septem leves, ut b, *ó, %, ni, t, n, -p. Poetae autem docent 
de qualibet harum classe, servato jam ordine, consonas 
prioris classis nobiliores esse, seu majoris potestatis consonis 
quibuscunque subsequentium classium. Et dicunt consonam 
f principem esse omnium consonarum seu reginam ; post 
veró ipsam aiunt tres molles prsecellere aliis subsequentibus 
omnibus omnium classium. Similiter duras excellere asperas, 
et fortes prsecellere levibus, omnium utpoté ignobilissimis 
et debilissimis. 

De Prosodid Hibernicd. 21 

7. An initial f never accords with anything but another 
f, and then the consonant to which it may be joined must 
also accord ; therefore p does not alliterate with f t>, nor 
with f c, nor with f-o, nor with fj;, but a simple f requires 
a simple p . So ft» demands another ft>, and so of the other 
combinations. Similarly f does not accord with cf after 
the particle aw, but in order to make an alliteration cf requires 
another cf , which the article ay\ requires occasionally, as we 
have said before, for example, an cfúit, An cfti§e*. 

8. We have seen elsewhere (see p. 52 of the Grammatica) 
that the soft\ consonants are three in number, namely, c, p, c ; 
that there are also three hard, namely 5, t>, t> ; that three 
sounds are reckoned rough, cti, en, pn (or p ) ; that five sounds 
or combinations are called strong, it, nn, ff, tn, ng ; and 
that seven sounds are light, o, t>, 5, rh, t, n, f. The poets, 
moreover, teach concerning each of these classes that, keeping 
the above order, those of the first class are " nobler " or of 
greater power than those of any of the subsequent classes. 
And they say also that the consonant f is the chief or queen 
of all the consonants, but that after this, the three soft con- 
sonants c, p, c, excel all others in the succeeding classes. 
Similarly they say the hard ones u, t>, 5, excel the rough 
(en, en, pn), and the strong ones the light — these last being 
the lowest and weakest of all. 

* That an eclipsed f (cf ) only alliterates with another Cf is therefore 
a third exception to the rule that neither aspiration nor eclipsis hinders 
an alliteration. 

f Dr. O' Donovan and other modern Irish grammarians — without 
understanding or trying to understand the point of view — have criticized 
and condemned the old native classification of the consonants as 
unscientific — especially with regard to the names soft and hard. 
O'Donovan and others say c, p, r are not " soft," but " hard," and 
that 5, b, -o are the " soft " consonants. I believe these moderns 
are wrong, and that our old authors were right, according to what 
they intended to convey ; c, p, t are mutes, non-sonant or " silent " 
consonants — therefore our ancestors called them 005, soft-sounding 
or silent consonants, b, x>, 5 are sonant, audible, loud, and therefore 
they are cf uait>, voiced or loud (hard) sounds : bos an( ^ cjiuai-ó 
expressed, I believe, not the less or greater compression of the organs 
of speech, but the " soft " or " hard," i.e., the " low " or " high " effect 
of the consonant sounds on the ear. 

22 De Prosodiá Hibevnicá. 

9. Nota tamen quod m raró nisi in fine voculae sit longa, 
ut in ZAxx\, mAtn ; imó raró hoc ipso effertur longé quia con- 
sonae fortes — maximé finales — sunt mediae quantitatis in 
pronunciatione, mediae inquam ut suprá inter longam et 
brevem. Revoca in mentem quod supra docuimus de 
quantitate syllabse, vulgó fine, quam dixi triplicem, nempé 
longam brevem, et mediam, vulgó ^at>a, seA^ et rneA-ooriAC ; 
hinc longa linea ponitur supra t»Af, |\óf, etc., sine qua forent 
breves, ut bap, pof, bof, supra quae nulla apponitur linea 
designans quantitatem longam vel mediam. Verum media 
quantitas denotata per lineam non adeo longam super imposi- 
tam medio quodam tractu effertur, non sicut longa vel brevis, 
sed brevius quam longa, et longius quam brevis, ut cÁinc, 
geÁlt, T)órin, f eÁn^ de quibus adhuc redibit sermo. 

De Prosodiá Hibernicá. 23 

9. Note, however, that rn is rarely long* at the end of a word, 
as in cAtn,f mA\r\ ; rarely, indeed, is this pronounced long, 
because the strong consonants, especially finals, are of medium 
length in pronunciation — I say medium, that is, as explained 
above, between long and short. Recall to mind what we 
said above about the quantity of a syllable commonly called 
fíne,t which I have said is threefold, namely, long, short, 
and medium, or in Irish ipAVA, 5eÁr\f, A^uy me^onAC Hence 
a long line is placed over bÁf, f ó r> etc -> without which the 
vowels would be short, as, bap, for, bof, over which no line 
is placed. The medium quantity, however, denoted by a 
superimposed line, not so long as in the former case, is expressed 
by a certain medium length of pronunciation — not so much 
long or short — but short rather than long, and long rather 
than short, as in cÁinc, jjeAtt, *oónn, f eÁng — of which enough 
for the present — the subject will recur. 

* The author means not so much the letter m itself, or even its sound, 
as a syllable ending in m, especially a monosyllable, as tyiatti, Am, torn, 
com, -]c. Vowels and syllables are said to have quantity, and to be 
long or short, but length is not generally ascribed to consonants. 

f CAffl is probably the word now pronounced and written Am=time, 
occasion. The u is probably radical, cf. cAm-AUl=a little time, a 
while. The z dropped out under the influence of the article — An tatti 
was thought to be for An c-Atn. Cf. Latin urbs, orbis for turbs torbis : 
cf. also Ir. mmif; for nmmifi, A-p An -oofuf for caji An x>o\Kvtf=x:A\\ An 
x>of;vif:=rt,hrough the doorway. 

rjAm is probably the same as the Welsh tarn, tamaid=a. small piece, 
a bit. The root is found also in Lat. tem-pus, tern-plum, and is no 
doubt identical with the tí/á — in rs/j,vc*)=I cut. 

J Ó Molloy and some other writers after him confounded the 
Irish ríneA-ó, fine, a stretching, extending, lengthening, with the 
borrowed word fine older riijne for O. Ir. rijne, i.e., the Latin signum, 
a sign or mark : rijne f. A-oA=the long sign or mark — not the long 
stretch or extension. Our author uses fine for quantity, whereas 
it is but the sign of the long quantity ; but he spells it correctly f ijne 
in subsequent chapters. The confounding of the two words was 

24 De Prosodiá Hibernicá. 

CAP. IV. [xvii.]. 
De Correspondentia et Extremis. 

1. Correspondentia vulgó comAx?o& duplex est, altera sana 
vulgó ftÁn, altera fracta vulgo t>|\ifte. Sana consistit in 
convenientia duarum vocularum in numero syllabarum, et 
quantitate vocalium et immixtione consonarum consimilium, 
sen ejusdem classis, aut defendentium sese ipsas juxta ordinem 
syllabarum ; praeterquam quod initiativa consona non 
necessarió debeat sic correspondere cum ulla initiativa alterius 
voculae ; nisi quando plures consonae quam duae simul con- 
fluunt in initio, tunc enim perinde est quae illarum corres- 
pondeat alteri in altera voculá, quia sufficiet si altera aliqua 
maximé subsequentium sic correspondeat alteri in altera 

2. Unde quantum ad tres consonas molles c, p, c corres- 
pondebunt cuicumque subsequenti, correspondentia etiam 
sana : Sic correspondent ^oc, f op, toe, sunt enim mono- 
syllabae, ejusdem quantitatis utpoté brevis, ejusdem in 
specie vocalis, ejusdemque classis omninó consonae extra 
initialem non excedentium duas, quia omnes sunt molles, 
adeoque inter se correspondent. 

De Prosodid Hibemicd. 25 


Of Assonance and of Terminations. 

1. Assonance or rime — which we call co\x\a\coa — is two- 
fold, the one p lAn, that is, whole or perfect, the other bpipce, 
that is, broken or imperfect. The perfect assonance consists 
in the agreement of two words in the number of their syllables,* 
in the quantity of the vowels, and in their union with similar 
consonants either of the same class, or having them in the 
order of their syllables ; except that an initial consonant 
need not necessarily correspond with the initial of another 
word ; unless the initial consonants of more than two words 
are compared ; then, indeed, it is equal which of them corres- 
ponds to another in another word, because it will suffice 
if the subsequent consonants — those following the vowels — 
correspond with each other. 

2. Hence as to the three soft consonants, c, p, c, they will 
correspond to any subsequent letter of the kind, and such 
correspondence will be perfect ; thus poc, -pop, toe, correspond 
or assonate, for they are monosyllables of the same quantity, 
namely, short, of the same kind of vowel, and the consonants 
are all of the same class [except one initial the p of pop — 
for p and t are of the same kind], and therefore as the final 
consonants are all soft, the words assonate or correspond in 
every respect. 

* This agreement in the number of syllables is only needed in certain 
kinds of metre — as in juonnAjvo, and the two varieties of jtArmui jeAcr ; 
obviously not in -oeitM-óe where the riming words must have an unequal 
number of syllables. Still even in x>eit)i-óe we must remember that 
the accented and riming syllables must be of the same quantity and 
equal in number — one syllable to one, two to two, etc. 

26 De Prosodid Hibernicd. 

3. Sic correspondent jat) et LA5 propter identitatem 
vocalium et quantitatis et classis consonarum, cujus sunt 
t> et 5. Similiter correspondent •oe^crhoit) et teAtnoi'ó et 
leAtipoiT), quia conveniunt in numero syllabarum, item in 
quantitate, item in vocalibus et consonis, extra primam 
ejusdem classis, v.g., in t>, hinc inde in c, t et p. Similiter 
per omnia correspondent bA|V[\ et ^Arm et bAtt et Atn et bAnj; ; 
item concordant cAob, Aox>, LA05, CAom, r-^ojx ; item con- 
cordant t)AoL, pAon, cAotn, nulla habita ratione principiantis 

4. Adverte ex consonis quasdam mutare suam potestatem 
et naturam ob consortium aliarum, de quibus postea redibit 

5. Correspondentia fracta est duarum inter se vocum 
convenientia in numero syllabarum et vocalium et quantitate, 
nulla habita ratione consonarum quoad speciem vel genus vel 
classem, sed ut non constent consonis ad sanam deservientibus 
correspondentiam. Hujusmodi autem correspondentia fracta 
tolerat ut vocularum sic correspondentium altera finiat in 
vocalem et altera in consonam : sic fracté correspondent bA 
et btAf, item cap et cIaóu, item aoi et Aoip, item btAoipj; et 
bAoif, etc. 

De Prosodiá Hibernicd. 27 

3. Thus -&ÓX) and Ldg correspond or assonate on account 
of the identity of the vowels and of their quantity, and of 
the class of consonants to which -o and 5 belong. Similarly 
correspond -oeACtrioit) (tithes) and te-AtnoTó (spread thou) 
and teAnpoiT) (he will follow) because they agree in the number 
of their syllables, in their vowels and consonants, and also 
in the final x) (which is the first in its class — the leves or 
light, -Ó, t>, 5, t, n, f, tri) as well as in the other consonants 
c, t, -p. In the same way, in all respects rime bA^ and 
5Atin and t>AtL and Am and DAng ; also caod, Aot), tAo£, 
cAotfi, f Aojt accord* or rime ; and so rime t>aoI, -pAon, cAomf — 
no notice being taken of the initial consonants. 

4. Notice that some consonants change their power and 
nature by their union with others (i.e., by eclipsis, aspiration, 
etc.) — about which more will be said later. 

5. The broken or imperfect rime is the agreement of two 
words in the number of their syllables, in the identity and 
quantity of their vowels, without reference to the consonants 
as to species, genus, or class — so long as they are not of the 
class that form a perfect rime or correspondence. This 
broken rime moreover allows that one of the words so agreeing 
may end in a vowel, and the other in a consonant ; thus not 
only do caf and cIacc, btAoifs and bAoif form an imperfect 
cotíiajvoa or rime, but also da and bAf , A01 and Aoif . 

* The author's concordant is, of course, here used in the general sense 
of " agree," " rime " — and not in the specific technical sense in which 
he used it in the previous chapter — that of alliterate, make an alliteration. 

f CAom in the more exact metres would not rime perfectly with 
•oAot or fAon, for the m is a strong or heavy consonant riming with 
It, nn, 113, and not with 1 and n which are both light : a better example 
would be -DAo-fi or caoji — for t, n, ft when medial or final are all light. 
But in the less exact ostACAf metres, the riming of f Aon and caoiti 
would be tolerated. 

28 De Prosodiá Hibernicá. 

6. Termini, de quibus supra, dividuntur in majus extremum 
et minus extremum — vulgó jurm et Aijvofurm. Suntque duae 
voculae, quarum una alteram unica tantum excedit syllaba, 
inter se tamen conveniant, ut dictum est de correspondentia — 
in numero syllabarum etiam, excepta prima excedentis, 
nulla habit a ratione alias primae syllabse qua posterior excedit 
primam. Excedens autem vocatur majus extremum, et est 
postrema vocula istius semimetri in quo admittitur. Exces- 
sum veró, seu vox pauciorum syllabarum appellata minus 
extremum est ultima dictio primi quarti cuj usque semimetri 
in quo admittitur ; adeoque haec vocatur jurm, ilia veró 
Ai|vp|\inn. Sit exemplum gtAf et r-on-Af, item cACAfi et 
cuAfc^cAT), item sAtfAi-o et pu-AfiAUAif. Ex quibus voculae 
gtAf tanquam minori extremo bene convenit fon^f tanquam 
majus extremum, quando rite ponuntur in metro, et sic de 

7. Dixi autem in quo admittitur, quia ut supra, etsi admit- 
tatur in genere metri vulgó "oeiDroe, nequaquam similiter 
in reliquis quinque supra enumeratis. Sit ergo exemplum 
hujus metri in quo admittitur : 

Ó5IÁÓ "oo t>i A5 TTluifve rhoifi, 
TIac ecus eiceAC 'tiA ti-onoif : 

Vocula enim rhóif\ est minus extremum, vocula veró ti-onóin 
est majus extremum. 

De Prosodiá Hibernicd. 29 

6. Terminations or points, spoken of above, are divided 
into the less and the greater — or, as they are called, the jurm 
and the Ái-fVOfurm. They are two words one of which exceeds 
the other by a single syllable, but let them agree — as has 
been said before about assonance — in the number of syllables, 
except the first syllable, no account in this respect being 
taken of the first syllable by which the second furm exceeds 
the first. The exceeding or longer point is called the greater 
or Ái|vofiintt, and is the last word of the semi-metre in which 
it occurs. But the exceeded point or the word of the fewer 
syllables called the lesser, is the last word of the first quartan 
of any semi-metre in which it occurs, and so this is called 
the f\wn or point, the other the Ái-|roturm or long point. Take 
for example gtAf and r-onAf, also z&ó&p and cua^zacat), also 
S-Ati-Aift and -ptMHAtiAif. Of these, with the word gt^f as 
minor point the word fotiAf as major point well agrees, when 
duly placed in a stanza — and so of the rest. 

7. I have said in whatever verse it is admitted, because, as 
said above, although admitted in the metre called •oeitii-oe, 
by no means is it equally admitted into the five* others 
above enumerated. Let this serve as an example of this 
metre : 

Ó5LÁ0 *oo tM 45 tTluifie tíiói^ 
V\At CCU5 eice^c 'n^ n-onóif : 

[Noble the servant of Mary, 

Never refused he to honour her !] 

Here rnóijt is the minor point, and ti-onoif* the major. 

* " Five others " : recté quatuor, for -oeibi-oe is one of the five metres 
enumerated in Cap. I., p. 4. 

30 De Prosodiá Hibernicá. 

Similiter fieri debet in secundo semimetro, per omnia suo 
modo, ut in sequenti integro videri est metro : 

Í1.A01 ccéx) if cjvi pidt) peio, 
As t\i§eA"ó "o'GitífAbecc, 
Sóifi|A 5L1C Apmt& av\ fumn, 
Cóifi|\ CdLttAij rhic Contntt ! 

vel sic : 
5 eu 5 01-Le ^Af t1t) ó tftoro, 

"b-á-pn uifjie ni t)-puAi^ AoncoiT) — 
"P-átfoitnín nÁ-|A ctAon te cjviot 
Ájvoóoitín fAOfi a -pinfio|A ! 

8. Unio vulgó tiAitne est duarum convenientia vocularum 
inter se, sicut dictum est de correspondentia, praeterquam 
quod non postulet vocales utrobique esse easdem, tametsi 
requirat ut haec interveniat inter subtilem et subtilem, item 
inter largam et largam, sic enim et non aliter uniuntur ut 
videri est in CAot et tnAot, item inter f ao^aíca et aottoaíca, 
item inter 05 et U5. Syllaba enim larga nequit uniri subtili 
ut us et tag. 

9. Verum si voculae non sint monosyllabae, sed poly- 
syllabse, suííiciet eas convenire in subtilitate vel largitate 
ultimarum syllabarum, ut a>oX)a et hwotiA, item -mtíie et •ooirhne, 
item o^rhAiite et -pe^n^oi^e. Si tamen in omnibus con- 
venient syllabis, nine inde utrobique largis et subtilibus, 
vel tan turn largis, vel tantum subtilibus, eo foret melius 
et gratiosius et dulcius. 

De Prosodiá Hibernicá. 31 

So it ought to be in the second semi-metre in all cases of 
this measure — as can be seen in the following complete 
stanza : 

Thrice three hundred and three score — 
Tale unheard by thee before — 
Feasted free in Calvach's hall 
Caring light what might befall !* 
Or thus : 
Youth who never flinched from fight — 
None could beat and none affright — 
Never bent the knee through fear — 
Poplar tall without compeer ! f 

8. Union which we call tiAitne} is an agreement of two 
words much like what has been said of correspondence or 
assonance, except that it does not require the same vowels 
always — although it demands that this agreement should 
be between slender and slender, and also between broad and 
broad, for thus and not otherwise are words united or matched ; 
as may be seen in cAot and tnAot, -pAojAtu^ and aottoaixa, 
also in 05 and V15 — but a broad syllable cannot be united 
or matched with a slender, as 115 and 145. 

9. But if the words are not monosyllables but polysyllables, 
it will suffice if they agree in the slenderness or breadth of 
their last syllables, as AtfoA and biotrtDA, also imfie and TJoirrme, 
also opmAitte and feArifvOi^e. If, however, they agree in 
all their syllables, first and last, everywhere, in broad and 
slender, or have only all broad or all slender, the verse will 
be all the better and the sweeter and the more graceful. 

* Literally : 

Nine hundred men and three score — 

When Elizabeth was reigning — 

Were the company of Calvach, son of Conall ó Molloy, 

The wise, strong-armed young chief of his land ! 
t Literally : 

Another young branch that never flinched from fight 

Over whom none ever gained triumph — 

The sheltering wall that never bent through fear — 

Lofty tree of noble ancestry ! 
t tUicne, from what is said here and in subsequent chapters, is 
clearly only another kind of assonance or but another name for it — 
especially for the initial and internal assonance. Although it resembles 
the word unto and is translated union, although even there was an 
old word xx&i for one whence -uaca — singular, and uacaxj the singular 
number, also, a small number, a few, it is doubtful if this is the original 
meaning and derivation of the word : it may mean a column, a support ; 
a word of equal sounds and syllables supporting or matching another. 

32 De Prosodiá Hibemicá. 

10. Caput vulgó ce-Arm appellatur monosyllaba vox quam 
semimetrum generis f e^-ón^ requirit in ultimo loco postremi 
quarti, cujusmodi sunt ionn in sequentis semimetri fine et 
piorm : 

Oi£f e Cát&oMp ciorm a cirivó \ 

1onrhtJin tirine 510*0 é iorm — ^x ^ 

Dj\A€AC Aige x\a CCU15 ccoi^eAt), 
CAtAó óispe^|\ úif\ nA ttpiorm. J 

11. tltALdrm ea vocatur vocula in initio primi quarti alicujus 
semimetri reperta, cui ut alia correspondeat aut correspon- 
dentiam faciat non est necessum, neque vitium si faciat ; 
unde si hujusmodi vocula nullatenus interveniat, eo longe 
majoris semimetrum est laudis, et vocatur inde ab Hibernis 
tÁitroetíriAtti, quod sonat perfecta compositio, perfectam 
intellige antonomasticé dictam. 

12. Aliud adhuc requiritur in metro et vocatur ^muf inque 
hoc consistit ut vocales sint ejusdem soni, ita ut vix non 
coincidat cum correspondentia fracta, de qua superius, 
praeterquam quod ^mtif semper et indispensabiliter in numero 
syllabarum postulat aequalitatem, ut hie : 

TTlÁf •OAonn^óc "oe-AttttA^ fwi X)Ár\ 
TTlÁf x>eAVb nó LAoCjVdCC nó tút, 
T)o nóf 5AÓ true foiling jvío§ 
tleic a gniorh m "ooiglit) *óún ! 

De Prosodiá Hibernicd. 33 

10. Ending, or ce-dnn* as it is called, is a monosyllabic 
word which a semi-metre of the kind called p e^tiA requires 
in the last place of the second or latter quartan, of which kind 
are ionn and bpiomi at the end of the following semi-metres : 

Oij^e Cacao1|\ cionn a cinró, etc. 
Cahir's heir, his people's chieftain, 
Loved by us and worthy all — 
Holds the flag of Erin fearless 
Leader peerless, praise not small ! 

11. "UpUrm [spearshaf t] f is the name given to the first 
word of any semi -metre — not being an unaccented mono- 
syllable — and it is not necessary that it should make a rime 
or have a word riming to it, neither is it a fault if it makes 
any such rime ; whence if a word of this kind does not occur 
(that is, if a semi-metre begin with an unaccented mono- 
syllable) the semi-metre is all the more to be praised, and 
such a te-At|\Atin (or half-verse) is called by the Irish tÁm- 
*oetm,Arh which signifies perfect composition — " perfect " being 
understood in a limited and figurative sense. [In the p e^-on-d 
verse above given oijpe is the upuvnti in the first semi-metre 
and is without a rime ; in the second bp^u^c is the up t^rm 
and rimes with c&t&ó in the next line]. 

12. Another thing required in some kinds of verse is called 
the Atntif, and this demands that the vowels should be of the 
same sound, and almost coincides with the imperfect asson- 
ance or rime, of which I have spoken already ; except that 
the Atrmp always and indispensably demands equality in the 
number of syllables, as here : 

If bounty be theme for bard 
If valour regard should win — 
Such traits in the son of my king 
I could sing whilst the world should spin ! 

* This term ceAnn is sometimes (as by O' Donovan and others) 
rendered head, and it is possible that ó Molloy may have understood 
it in that sense ; but the Latin caput and its French representative chef 
also mean end, and the Irish ceAnn very often means end as in ceAnn 
mior-A (the end of a month) ceAnn btiA-ónA, x>Á ceAnn x>e iriAroe=:the two 
ends of a stick. To call the last word of a couplet the " head " is, I think, 
an abuse of language, and I have translated the word ceAnn therefore 
in this sense ending. I take it to be simply a monosyllabic ftmn. 

1 11 fit Ann in this place is clearly but a fanciful name for the first or 
leading word of a teAtfiAnn, and is from the same figure by which 
the end of a line is called its point. Perhaps both were originally used 
only of the " shafts " and " points " of a line of satire. Cf. the English 
expressions, " shafts of wit," " shafts of ridicule," " a fool's bolt." 

34 De Prosodid Hibernicd. 

13. Sunt qui pro Amur- faciendo ponunt a ad correspon- 
dendum vocali e. Sed raró : nee hos in hoc laudo nee 
imitandos puto, quia non bene consonant utpoté subtilis 
cum larga, sed faciunt dissonant i am. Oi, in syllaba brevi 
facit Atrmp seu consonat cum a\ similiter brevi, ut in c^oit 
et -puvit consonant nam que utrinque subtiliter et sonat 01 
quod ai ; loco hujusmodi Atnuf bene potest deservire corres- 

CAP. V. [xviii.]. 
De metro Hibevnis •oeifti , óe. 

1. Hujus generis semimetrum requirit ut primum ejus 
quartum finiat in voculam quae sit minus extremum vulgó 
jurm. Secundum veró ejusdem semimetri quartum desinat 
in alteram voculam quae sit majus extremum. 

2. Quodlibet quartum cui usque semimetri postulat constare 
septem syllabis, non pluribus nee paucioribus in scansione, 
ut ante dictum est, sive inter veniat elisio alicujus vocalis 
sive non. 

3. Primum hujus generis semimetrum, vulgó féotd'ó, ultra 
minus et majus extremum, et dictum in quartis syllabarum 
numerum, adhuc requirit in quolibet quartorum, illam de 
qua supra dixi concordiam, vel propriam vel similitudinariam 
ita ut in utroque quarto primi semimetri ad minus interveniat 
similitudinaria, elegantior autem foret propria, sed propria 
semper citra licentiam requiritur in quartis secundi semi- 
metri, scilicet inter ultimam quarti voculam et aliam imme- 
diate praecedentem, vel quasi immediate et valde vicinam. 

De Prosodid Hibernicá. 35 

13. Some in making an Atrm-p put an a to rime with e 
but rarely ; nor do I commend these, nor think they should 
be imitated, because these vowels do not rightly harmonize* 
one being slender, the other broad — in fact, they make 
rather a dissonance. Oi in a short syllable makes an -Atrmf 
with A1 also short, as in cj\oit and ptdit, which rime because 
the broad vowel in each is modified by a slender, and in fact 
01 short sounds as ai. For this kind of .arrmr an assonance 
can well stand. 


Of the Metre called by the Irish •oeifti'óe. 

1. A couplet of this kind requires that its first line should 
end in a word which is the minor point or termination called 
f\irm. But the second line of the same couplet must end 
in a word which is called the major point or Áijvotvirm. 

2. Any line of either te^t^rm (or couplet) requires seven 
syllables, neither more nor fewer in scansion — whether, as 
has been said before, the elision of any vowel occurs or not. 

3. The first semi-metre of this kind — the leading or f eoUvo — 
besides the minor and major point, and the due number 
of syllables, requires further in each line that concord or 
alliteration I have spoken of above — either the true species 
or the apparent, so that in each line of the first couplet at 
least the weak alliteration should occur, but the proper or 
true species would be more elegant ; but the true always 
without any licence is required in both lines of the second 
semi-metre, that is, between the last word in the line and 
another immediately preceding, or almost immediately and 
very near. 

* Harmonize : the author's consonant is a very appropriate word 
here for assonate, agree in sounds ; this agreement or harmony of 
sounds is, indeed, called consonantia as often as assonantia by 
continental writers on the subject. 

36 De Prosodiá Hibernicá. 

4. Secundum veró semimetrum plus adhuc exposcit, 
nempé ut nullum sit verbum vel nomen in ultimo ejus quarto, 
praeter majus extremum, cui non correspondeat aliud in 
anteriori ejusdem quarto ; quia nullum debet esse in primo 
quarto hujusmodi secundi semimetri, vel verbum vel nomen 
(si exceperis ufvtAnn et minus extremum) cui non corres- 
pondeat alterum in sequenti quarto ejusdem. Dixi nomen 
vel verbum, quia inter haec et similes voculas non computantur 
adverbia vel articuli, ut alias dixi. Sit exemplum : 

Ó5LÁÓ "oo bi A5 ÍTIuife tfióif, 
TlAC ecus eiceAc x\& n-onói|\, 
teif tiÁ|i b'Áit T)on tnte dan, 
AmÁm Atz tTluif\e mÁtA^. 

5. Ubi insuper necessum est ut concordiae ultimi semi- 
metri sint propria^ et non similitudinariae. Declaro omnia : 
Vides si numeraveris quodlibet quartum septem constare 
syllabis, non amplius, facta scansione ut in primo quarto 
allati jam semimetri, in quo intervenit elisio inter ui et 45 : 
vides prseterea minus extremum quod exceditur a majori 
et cum eadem quantitate ejusdemque speciei syllabse con- 
tineatur in majori extremo ; nam sic \y\ó\\\ comprehenditur 
in n-onói|\. Vides prseterea qualiter in eodem quarto reperitur 
concordia etiam propria inter 1Tluif\e et rhóifi, utpote con- 
cordantibus in initiativa consona. 

De Prosodiá Hibernicá. 37 

4. But the second semi-metre requires further that there 
should be no (important word) verb or noun in the last 
quartan — except the major point* — to which another in 
the previous line does not rime ; for there should be no 
word in the first quartan of this second semi-metre — whether 
verb or noun — if we except the uj\tdrm or leading word and 
the minor point — to which another does not correspond in 
the second of the same. I have said noun or verb because 
amongst such and similar words we do not count adverbs 
or articles, as I have said elsewhere. For example : 

ÓgtÁó *oo oi as tTtatfie rhóif\ 
Tl.dc ccug eice.dc 'via ti-onoi^, 
1eif nÁf\ u'-Áil T)on uite o^n, 
ArhÁin act: tttuit\e mÁtAj\!f 

5. Here it is necessary, moreover, that the concords (or 
alliterations) of the last semi-metre should be the true ones 
and not the apparent. I will now show that this example 
has all the requisites (of Irish •oeiuroe). You see, if you have 
counted, that each quartan consists of seven syllables, not 
more, as when scansion is made in the first line, elision occurs 
between oi and ax; : you see also the minor point (tfiói^) 
exceeded by the major point fi-onóif, and with the same 
quantity and the same kind of syllable, the firm may be 
contained in the ,dijvo-]\irm — thus rhóif is comprehended in 

* It is curious that the author for the moment does not seem to 
consider that the -junn rimes or " corresponds " with the Áifro--|tinn — 
whereas this correspondence (in the accented syllables) is one of the 
chief characteristics of this metre. It is true, however, that whilst 
junn and Ai-p-ojiinn were used for the final rimes, another name viAicne 
was used for the internal rimes, as b'Áit and ArriAin, uite and ttluifie. 
Here, again, not the nature of the thing but the position makes the 

f Literally : 

Noble the servant of Mary — 
Never refused he to honour her — 
Count all the fair race of women, 
None loved he like unto Mary I 

38 De Prosodid Hibernicd. 

6. Vides praeterea qualiter in eodem quarto reperitur 
concordia etiam propria inter tntnj\e et rhoi-|A, utpote con- 
cordantibus in initiativa consona. Porró vides qualiter finalis 
consona minoris extremi habet similiter finalem consonam 
majoris extremi, ejusdem secum classis, nempé consonam 
levem, cujusmodi est ft utrobique. Item vides in secundo 
quarto prioris semimetri qualiter interveniat concordia inter 
eiue-Aó et ti-onói|A, quia incipiunt utrinque a vocali, nam ad \\ 
non attenditur, quia non est possessiva, sed accidentaria. 
Vides praeterea in primo quarto secundi semimetri concordiam 
esse inter bti'Áit et tnte, incipiunt enim utrinque a vocalibus, 
nee ofneit u in ti'Áit quia accidens est et non possessivum 
istius voculse. 

7. Rursus vides minus extremum ejusdem nempe x>ax\ 
comprehendi in majori subsequentis, nempé mátAft, in quibus 
convenitur etiam in consonis finalibus, nempé n et ^ ^^ 
sunt leves ejusdemque classis. Adhuc vides Ait et &m Áin 
habere unionem ut supra die tarn, eamque insuper inter venire 
inter 1Tlui|\e et tnte. Item finalem consonam majoris et 
minoris extremi, nempé voculae oAn et mÁt&^ esse ejusdem 
classis utpoté utrinque levis. Caeterum omnia requisita ad 
hoc genus metri ibi reperies. 

De Prosodid Hibernicd. 39 

6. You see also how in the same line there is a concord 
or alliteration — even the true one — between triune and 
mói^, as they agree in their initial consonant. Further, 
you see how the final consonant of the minor point 
is similar to that of the major — they are of the same 
class, namely light, and consist of the letter fi in both 
cases. There is also to be seen in the second line of 
the first semi-metre how there occurs a concord between 
eice.dc and n-onói^, because they both begin with a vowel — 
for one does not consider the n, as it is not radical, but 
only accidental. Then, again, in the first line of the second 
semi-metre there is a concord between ft '-art and uite, for 
they both have vowels for initials, nor does the t> in Io'áiI 
hinder this because it is an accident and not a radical part of 
the word. 

7. Again you see the minor point of the same line — namely 
DAn* comprehended — so to speak — in the subsequent major, 
which again agree in final consonants, as n and jt are both 
light and of the same class. Further, you notice that AM 
and AtfiÁin have a union or rime as above explained ; another 
also there is between triune and ttite. Also you see that 
the final consonants of the major and minor points, i.e., of 
the words X)&\\ and tnÁú-dft are of the same class, namely, 
both light. You therefore find in the verse all that is 
demanded for this kind of metre. 

* This curious word bAn (-oon tube bAn) does not appear to be used 
at present, nor do I call to mind seeing it in this sense in any other 
author. It is not the word beAn, though, no doubt, connected with 
it, beAn itself making b<\n in the genitive plural, and as a prefix 
bAn-ofiAoi, bAn pi Ait, -jc. Here it is a dative without inflection of some 
nominative bAn which was a collective (prob. fern, with gen. bAn a) 
like bAnr-fiAcc, An cméAt bAn-oA=womankind, the female race. T)on 
•uite bAn=of all womankind. 

f Notice mÁúA-p used here as a qualifying genitive I for the particular 
phrase TTluifie rnÁtAfi we now mostly use muijie rriÁÍAift or tnÁtAifi, 
putting the second word in apposition to the first, here in the nominative. 
But we have many phrases like mmjie rnÁiA-fi : cf. niACAom mnA, céite 
fifi, mAtjiAc fMfi, bAmu-peAbAC pift, njeA-pnA eArpui5, etc. 

40 De Prosodiá Hibernicd. 

8. Ut hie : 

U15 W tQAt^X" UA1 F oite, 
C15 Via cúigeAjA cUvooij\e, 
U15 Via *óír 05 'ha twine, 
C15 Aftír 'wa TtufU4roe ! 
Item hie : 
5t-Aine nó các tú triAfi tfiAt, 
ÍAige a uc^acc nó UijuaL, 
SuttÁitce 5-áti óéitn A|\ cut 
U^én |\e mitiÁitce a *óeACtú ! 
Latiné : 
Nec te candidior Tyrel, nec firmior extat, 
Cui comes ut virtus nescia fama mori. 

Sicut et in omni alio metro hujus generis observabis. 

9. Difficile quidem factu apparet hoc metri genus verúm 
difncilius creditu quod superius allatum ÍU01 ccé-o, etc., 
refert ; verissimum tamen, cujus ipse oculares vidi et audivi 
testes fide dignissimos : nempé quod Carolus Conalli filius 
Molloyorum princeps, avus illustrissimi nunc viventis, vastato 
Hibernise regno fame, flammá, ferro, sub Elizabetha regina 
in summis annonse penuriis ; invitatos a se pro Christi Nata- 
litiis, per dies duodecim tractaverit nongentos sexaginta 
homines in domo propria. 

De Prosodiá Hibernicá. 41 

8. Or as here follows : 
U15 'tiA ceAt-p^n uai|\ 01 te, 

C15 Via óúigeA^ ctá'óoi^e, 
U15 'nA "óíf, C15 Via *óuine, 
U15 ^|\íf Via ftuf\uróe ! 
Here's the rogue with others three — 
Next the knave with four you'll see, 
Now again he'll come with one, 
Or as Rury quite alone ! 

Or like this : 
^tAine nó các tx\ rriAtt tjUAt, 
tAi^e a ccfÁcc nó Uij\iAt,* s^n óéim A-p cut 
Ufén fie mio Alice a "óeAótú ! 
Nobler than all as a chief, f 
Tyrell great beyond belief — 
Strong 'gainst wrong thy noble mind, 
Worth that never fell behind ! 
Thus, also, you will find every other poem of this metre. 

9. Difficult to write, indeed, appears this kind of verse, 
but more difficult still to believe is the fact mentioned above 
in the example given at p. 30 (" Thrice three hundred and 
three score "). But it is perfectly true, and of the fact I have 
myself seen and heard witnesses most worthy of belief : 
namely, that under the rule of Queen Elizabeth, whilst the 
kingdom of Ireland was in many places devastated with 
famine, fire, and sword, and in dire straits owing to the 
scarcity of provisions, Calvach, son of Conall, prince of the 
ó Molloys, grandfather of the most illustrious present chief, 
having invited to his house nine hundred and sixty persons of 
all ranks for the feast of Christmas, freely entertained them 
there for the space of twelve days. 

*TheCittiAt mentioned here is, no doubt, Father Patrick Tyrell, 
ex-dimnitor of the Franciscan Order, and one of the censors to whom 
Fr. ó Molloy's Grammatica was submitted. He was probably a 
native of Meath — at least of the diocese — like the author himself. 
The rest of this poem, which the author gave as an illustration of 
■oeiSi"oe, is given at the end of this book. [The name rjifUAl is also 
spelt CiftiAtt and C-fHAlt : cf. OeA-pnA An CjiiaíIaij or UeAÍAc &r\ 
C-pi4llAi3==Tyreirs Pass in Westmeath.] 

f As a chief, i.e., as guardian or superior. 

42 De Prosodiá Hibernicd. 

CAP. VI. [xix.]. 
De metri genere Hibernis Seu -on a. 

1. Seudnae genus requirit octo syllabas in primo quarto 
utriusque semimetri ; in reliquis veró septem non plures nee 
pauciores in scansione, sive elisio interveniat sive non. 

2. Porró petit ut primum turn primi turn secundi semi- 
metri quartum vocabulum finale habeat duarum praecisé 
syllabarum, reliqua veró quarta finiant in voculam unius 
syllabae : ut — 

ttlAij^e geAt -p,A eiciD e&lA 1 . 

'O^eioL |\e bpeAjA cc^eAjA rém. j 

Item : 
Hi via trout Ati |\1 *oo foijne 


1. corhAT». 

T)o ní úfi T)oti coirmte ó|\íon. 

Sunt autem corhAT* non r-eotA*, id est secundum semimetrum 
non primum. i 

3. Vocula monosyllaba qua secundum finitur quartum 
ab Hibernis vocatur bjAÁije ante quam immediate praecedere 
debet alia vocula bissyllaba. Supportatur tamen aliquando 
interj acere inter ipsas lAfirnbetifvtA seú articulus adverbialis 
seú adverbium, quale tunc etiam vocatur b^Áije. Verúm 
etsi supportetur reputatur pro vitio in hoc rigorosi carminis 
genere, ut hie : 

Cuimmg 50 b-ptnt a jú at) ^ottA 

£tllt 11A CCf\í cCottA Af "00 Cut ! 

— ubi inter cCo Ua et out jacet adverbium aj\ : de *oo enim 
nihil curatur quia non est adverbium. 

De Prosodiá Hibernicá. 43 


Of the kind of metre called SetrotiA. 

1. The metre called SetronA requires eight syllables in the 
first line of each semi-metre, but in the rest seven, not more 
nor less in scansion, whether an elision occurs or not. 

2. It requires further that the first line of each couplet 
should have its final word of two syllables precisely, but the 
other lines should end in a word of one syllable, as : 

rrUi§j\e seAt pA eicib eAtA I Second couplet of a 

T)'eicit |\é DpeA^ zz^eA^A céro. y feuwiA verse. 
[Literally : A white salmon darts under the wings of a swan 
That flew at the man with the spear.] 
Or this : 
ftí tiA rvoút ah -pi *oo foijne, \ Second couplet of a 
T)o ní úf\ *ooti comnte c^íon. j f etroriA verse. 

[Literally : The King of Creation he who made it — 

He that makes green once more the old withered stubble.] 
These, however, are all examples of the corhA-o or closing 
couplet — not of the f eotA-ó or leading distich. 

3. The monosyllabic word in which the second quartan 
ends is called by the Irish b|\Ái£e* before which there should 
come immediately a word of two syllables. It is allowed, 
however, that between these there may occur an lAftnbeufaA, 
that is an adverb or adverbial particle, which does not affect 
the bfvÁrse or " neck." But although allowed it is considered 
a fault in this strict species of verse, as here : 

Ctnriinrg 50 bpmt a fvT a*o f\ottA 

"ptut ha cc|ú cCoLLa A|\ *oo óút ! 

[Think, O king, that in thy pulses 

Blood of three great Collas runs !] 
— where between cCo Ua and cut lies the adverb aj\, for of 
•oo no notice is taken, as it is not an adverb. 

* This word b|tÁi je — lit. neck or throat — is merely another name 
for the monosyllabic ending of a couplet in peA-ona or in [the true] 
fiAnntnjeAcc beAj, another name for the ceAnn or ending explained 
in Chapter IV., looked at from another point of view. What in 
one sense is the ending of a couplet or verse, may also be looked on 
as the neck or connecting link between two couplets or two verses. 
The term in Chapter VIII. is spelt djiájai-o, but this is really an 
oblique case of the word spelt bjiÁtjA or bfiAije in the nominative. It 
agrees with the Welsh brefant and the Latin vSrdg-o s not with gurges, 
with which it has been generally compared. 

44 De Prosodiá Hibemicá. 

4. Nihilominus si hujusmodi adverbium eclipsatur seu 
eliditur seu mergi contingat in monosyllabam, tunc nullius 
vitii erit nota : ut hic — 

Acc lonstiA'ó a^ poj\Aiu £oriióifv 
ClOt) pÁ CC10b]AAt) onóifi t/Aot). 

Item : 
lotnftA pie 'já ttpuil, Ai^e 
'5 Á PS 6 Á cctnj oroe "o'eoin ! 

— ubi adverbium t>o eliditur per sequentem monosyllabam 
Ao-ó in primo semimetro. Item per monosyllabam eoin 
in secundo semimetro — ut alias diximus de mersione vulgó 

5. Hujus seudnae quartum duas continebit voculas ad 
minus quarum neutra sit adverbium vel articulus, earumque 
una sit finalis et altera in vicinio et ambae coéant ut supra 
dixi Cap. praecedente, ita ut ambae incipiant a vocali aliqua, 
vel certe ab eadem in specie consonante, ut bene concordent, 
ut dictum est Cap. XVI., quo te remitto, ut : 

Uj\í ^Á^tA Ay gnÁtAó 't\a •oútia'ó — 

— ubi sÁptA et gtiAtAó in primo quarto primi semimetri 
concordant seu conveniunt extra voculam nnalern in consona 
initiativa 5, quod sumcit ad ejus concordiam, tametsi non 
requiratur quin suffecisset hujusmodi concordiam intervenire 
inter ultimam voculam et aliam in anteriori, ut alias docuimus 
cap. praecedente ; ubi adverte a nnalern in ^ÁptA per a initia- 
tivam in Ay elidi et ex utraque fieri unam syllabam in 

De Prosodiá Hibernicá. 45 

4. Nevertheless, if an adverb of this kind is eclipsed or 
elided, or happens to be merged in a monosyllable, then it 
will be no mark of a fault, as here : 

Atz lon^tiA* a% piofVAit) porhói^ 
00*6 pÁ cciob|v<vó onóijA 'o'Ao'ó ! 
[Wonder great with Fovor's soldiers 
Why such glory's given to Aodh !] 

Also : 
lonró-A pile 5A bptut Ai%e 
'5Á pi§e a CC1115 oi"oe •o'eom ! 

[Many a bard of those around him 
Now is sounding Eoin's high praise !] 

— where the particle t>o is elided by the following mono- 
syllable Ao"ó in the first semi-metre. So to the monosyllable 
Coin the same thing happens in the second couplet, as we 
have said elsewhere in speaking of daca-o or sinking. 

5. A quartan of this fetrotiA will contain two words at 
least (neither of them being an adverb or article) of which 
one is to be a final and the other close to it, and both of 
them, as I have said in a preceding chapter, should be such 
that they begin with a vowel or with the same consonant, 
in order that they may alliterate truly, as said before in 
Chap. 16 [Chap. III. of the Prosody] to which I refer you. 
As for example : 

Uflí s&ptA Af 511ÁCAÓ 'tiA "outi^vo— - 

[Three sounds that sound in that fortress] — 

— where ^Á^tA and ^ÁtAó in the first line of the first couplet 
alliterate or agree in their initial consonants, outside the 
final word — which suffices for the alliteration although it 
may not be required — for truly it would have been better if 
the alliteration occurred between the last word and another 
preceding word, as I have shown in a previous chapter. 
Notice, also, that the final a of gAfvtA is elided by the first 
vowel of Af , and from both there is made but one syllable 
in scansion. 

46 De Prosodiá Hibemicd. 

Cftí 5Á|\tA -Af ^nÁtAó 'nA •óúriA'ó, 
'Oún.A'ó Ájvo a n-it)te|\ coptn — 

5á1|\ té"o if 5A1 |A jeirhe^t ngoftn ! 

6. Rursus hoc quartum finit in dictionem -ourixvo utpote 
bissyllabam, adeoque hoc quartum habet ex se omnia neces- 
saria ad primam medietatem primi semimetri. Nunc ergo 
eamus ad alteram quae sic habet : TmriA-o Afro a v.-mte\\ co^m, 
in qua vides septem syllabas, ultimam veró voculam esse 
monosyllabam necnon concordiam dari inter ájvo et n-iDtejA, 
eo quod incipiant utrinque ab a et 1 vocalibus, n quippe in 
n-it)teji non est propria vel possessiva, sed accidentaria et 
mere casualis, qualis non impedit neque deservit ad concor- 
diam, ut alibi diximus. Nee aliud requiritur hie ad primum 
seudnae semimetrum. 

7. Transeamus modo ad primum secundi semimetri quar- 
tum quod sic habet : 

5^1 f\ nA fcéT) a ti*oiit ti^ troei5peAf\ — 

ubi observa octo tantum intervenire syllabas et ultimam 
voculam esse bissyllabam, et concordiam reperire inter rroÁit 
et rroeij-pe-A^, non propter n utrobique repertam quae nullius 
est positiva vel propria, utpote utrique adventitia, sed propter 
X) quae et initialis est et proprié positiva ambarum vocularum. 
Rursus observatur concordia in hoc quarto inter ultimam 
voculam et aliam anteriorem, quod semper est necessum, 
sicuti etiam in ultimo quarto secundi semimetri, ut alibi 

De Prosodiá Hibernicd. 47 

Cftí j;Á\\tA Af ^nÁtAó 'tiA •óútia'O, etc. 
Three sounds e'er heard in his fortress — 
That fortress where feasting reigns — 
The neighing of noblemen's horses, 
With harp-play — and clank of chains ! 

6. x\gain, this first line finishes with the word -úún&i), that 
is, a dissyllable, and so this quartan has in itself all the 
conditions necessary for the first half of the first couplet. 
Now let us go to the second line, which reads thus : 

TXinAT) ájvo a n-it>tef coj\m : 

— in which you see seven syllables, the last being now a 
monosyllabic word ; there is also a concord or initial agreement 
between a^o and n-iDtefv inasmuch as they begin respectively 
with the vowels a and 1 ; the n in n-iDte|\, of course, is not 
radical or fundamental, but accidental and merely casual, 
such as does not hinder or make an alliteration, as we have 
said before. 

7. Let us now pass to the first line of the second couplet, 
which is as follows : 

5Aif n-A fcéx) a nT)Ait n4 troeigfreAf. 

Here observe that there are eight syllables, and only eight — 
the last word being again a dissyllable, an alliteration being 
found between troAit and rroeijire-djA not on account of the 
n found in each, which is not radical in either word and only 
adventitious, but on account of the initial t> which is funda- 
mental in both words. In the next place, alliteration in 
this line is observed between the last word and another near 
it, which is always necessary, as is also the case in the last 
line of the second couplet, as we have already explained. 

48 De Prosodiá Hibernicd. 

8. Nunc pergamus ad ultimum hujus semimetri, quod 
sic habet : 

5^1^ ten if sAij\ jeirheAt rtgoftn ! 

in quo vides septem tan turn dari syllabas, ultimam praeterea 
dictionem esse monosyllabam omninó convenientem in 
quantitate et sono, imó et finali consona levi ultimae voculae 
monosyllabae primi semimetri, nempé coftn. Quae enim 
major convenientia quam ilia reperta inter co^tn et nsojun, 
abstinendo ab identitate ? 

9. Adhuc vides ut requiritur dari concordiam non simili- 
tudinariam, sed propriam inter flnalem voculam n^o^m et 
penultimam geirhe-AL. Incipiunt enim a consona 5 tanquam 
propria et positiva utriusque cum n propria non sit sed 
adventitia ultimae voculae. Praeterea vides hie dari corres- 
pondentiam ut requiri docuimus Cap. XVI., inter rvoÁit 
primi quarti hujus posterioris semimetri et s-áift secundi 
quarti ejusdem, necnon inter rfoeigfre-dp et geirhe^t. Item 
inter f céx> et té*o in iisdem. Et sic omnia necessaria ad hoc 
genus metri jam enumerat interveniunt. 

De Prosodid Hibernicd. 49 

8. Let us now go to the last line of this second couplet, 
which is as follows : 

5-Áif\ téx> if 5^1|A jemie^t tigofAm ! 

in which you see but seven syllables, the last, moreover, a 
monosyllable riming thoroughly in sound and quantity and 
even in its final light* consonant with the monosyllabic final 
of the first couplet — namely cofun. For what greater agree- 
ment or assonance could there be than that found between 
cofAtn and nsotvm, short of absolute identity ? f 

9. You see also, as is required, there is an alliteration — 
not the apparent but the real — between the final word n^of tn 
and that preceding it, genii eat. For they begin with the 
consonant 5 as a proper and radical letter in each, since the 
n is not proper but adventitious in the last word. Besides 
this you observe there is an uAitne or internal correspondence, 
as we showed in Chapter XVI. is necessary, between tvoAit 
in the first line of the second couplet and 5^1 f\ in the second 
line of the same, as well as an Amuy between n-oeigpeAfv 
and $eitfie.<a p also an uAitne between r-rjetD and téx> in the 
same lines. And thus all the conditions necessary for this 
metre already enumerated are found in this verse. 

* Recté strong or heavy — for m is rightly reckoned with it, nn, ft ft 
and nz; — the consonce fortes — but aspirated m (i.e., rh or mh) is among 
the leves or light, t>, -ó, 5, rri, t, n, ft. See Gram. Lat. Hib. p. 160. 

f Yet this total agreement or identity of vowels and consonants seen 
in c-oftm and 5-oftm, though occasionally found in Irish verse, is not 
the usual — not the corhAft-oA which demanded identity of vowels, but 
rather variety of consonants, these latter, however (as explained by 
the author in Chapter IV.) to be of the same class. This, indeed, is 
the real difference between the Irish corriAffoA or assonance, and the 
English so-called " perfect " rime. The delicate-eared Celt was not 
satisfied with mere identity of vowels and consonants — which to 
him would be monotonous, but sought rather variety in the consonants 
that followed the vowels. 

50 De Prosodiá Hibernicá. 

10. Genus ipsum etsi sit difficile est elegantissimum, cujus 
gratia annexam addo compositionem, Latiné etiam explica- 
tam, esto nee cum tanta emphasi nee cum aequali vel nervo 
vel sue co : 

A ut.Áit tiA mvLóíó a pltiijA pumcécc, 
A ia^Um-o 5tó|\rhAin £inró $aIL, 
"Ó^oid Af cai|\i*ó t>fi^5tti^A t)|M$T)e 
mite bti^jAn 'f ctnite tA\X ! 
Latiné : 
Flos procerum, Fingalle Comes, Plunchette planeta, 
Mille tibi post hoc, Brigida iesta paret ! 

De Prosodid Hibernicd. 51 

10. The metre itself, though difficult, is very elegant, for 
which reason I give the annexed composition,* explained also 
in Latin, though not with such force, nor with the same 
strength or savour : 

A btAit r\A tnuA'ó a ptúifi pumcécc, etc. 


O Fairest flower of the Plunkets 

Worthy of honour and love — 
Thine thro' the prayer of St. Brigit 

Be ages of glory above ! 

* The rest of this Irish poem is given at the end of this book — see 
p. 109. It was addressed apparently to the young Earl of Fingall, 
then in Rome, probably for his education. The Blessed Oliver Plunket, 
Archbishop of Armagh, who was put to death for the faith in London 
in 1681 — only four years after Fr. ó Molloy's little work appeared 
in Rome — was a member of this noble family. 

The Irish verse given above is a very good example of the metre 
called féA-onA. The first and third lines have eight syllables each, 
the others seven each. A ptihji ptuncérc shows alliteration in the 
first line, jlojirhAijt, jaII in the second, ttpiA^rnAy "Ojiij-oe in the 
third, cuitle, rAtt in the last. The first and third end in a dissyllable, 
the second and fourth in a monosyllable. O-juj-oe has its Arn-up in 
mite, bfiiAjmAfi an d ctiifr.-p their uAitne (or internal rimes) respectively 
in btiA^An and ttntte. The English version given above is not meant 
to show all the points of the Irish verse — giving but the ideas in the 
most general way and in some sort of metrical form. 

52 De Prosodiá Hibemicá. 

CAP. VII. (xx.) 

De Metro magno et parvo 

necnon CAr-DAi^n. 

1. Numeratis jam duobus metrorum generibus oportet 
ad reliqua tria, ut prsemisimus, passum facere, quibus et 
primo generi commune est constare quatuor quartis, et 
singulis horum similiter constare septem tantum syllabis, 
ut dictum est, in quo et in aliis a reliquis differt f eu*on.A. 

2. Metrum itaque magnum, vulgo iwmoigeAcro rhó|\ ultra 
septem syllabas in singulis quartorum necessarió requirit 
finalem cuj usque quarti dictionem esse monosyllabam, hoc 
est unius tantúm syllabae. Prseterea necessum est interveniat 
unio de qua~supra, inter finales dictiones duas primi nimirum 
et secundi semimetri ultimas, quae utique debent uniri, seú 
unionem vulgó tj^itne habere inter se, verum non requiritur 
unionem intervenire inter reliqua extrema, tametsi postuletur 
quod interveniat inter duas voculas, quarum una sit in 
primo quarto, altera in secundo primi semimetri ; ex quibus 
prima vocula debet esse penultima vel quasi penultima 
ejusdem primi, quae cum corresponsali in secundo quarto 
ut dictum est convenire debet in numero syllabarum, in 
quantitate, in vocalibus, seu subtilibus seu largis, earumve 
sono. Cseterum non est necesse ut singula voculae primi 
quarti primi semimetri perfecte concordent cum suis 
corresponsalibus in secundo quarto ejusdem semimetri. 

De Prosodiá Hibernicá. 53 


Of Great and Little Versification 
also of GAfOAijui. 

1. From the two kinds of metre already described we 
must pass to the other three we have mentioned, in which 
also in common with the first kind (-oeiDi-oe) there are four 
quartans and seven syllables in each quartan — in which 
latter particular as well as in some others, f qax>x\a differs, 
as we have said, from all others. 

2. The metre called great versification* or jwmoise-dct) 
rhóp, besides the seven syllables in each line, requires also 
that the last word of each line should be a monosyllable 
or word of one syllable. Besides this, it is necessary that a 
union or uAitne (of which see above) should be made between 
the two final words f — that is, the last word of each semi-metre ; 
these should assonate or rime, but it is not necessary that 
the final words of the other lines should rime. It is demanded, 
however, that, there should be a correspondence or rime 
between two other words in each couplet, of which one 
is to be in the first line of the first couplet and a corresponding 
one in the second line ; and of these the word in the first line 
should be the last but one, or nearly so, of that line, and should 
agree with its riming word of the next in number of syllables, 
in quantity, in vowels, whether slender or broad, and in the 
sound of the vowels. As to the rest, it is not necessary 
that each word in the first line of the first semi-metre should 
have perfect correspondence with each word in the second 

* There is some reason for thinking that in naming this metre there 
has been an error which has been lightly passed on — that, in fact, this 
(with the monosyllabic ending) is the true fiAtittui^eAct be-Aj;, and 
that with the dissyllabic ending is the true -jiAnnuijeAcc móft. See 
my SeAcc SÁ-ffóÁtiCA ^Ae-óitse for a discussion of the point (Sealy, 
Bryers and Walker, Dublin). 

fThis shows that as explained in Chapter IV. uAicne is really but 
another name for coTriAfvoA or assonance, only given rather to the 
internal rimes than to the final ones. The riming of a final with an 
internal word in a succeeding line is an Amur- , and sometimes this term 
is given to the internal riming word itself. 

54 De Prosodiá Hibernicá. 

3. Sit exemplum : 

*0eAl5 AtÁloi-ó ot|táf Cai-ó5, 

T)Á|A n-Ancf\Átoiti> cocca An CU1L5, 
C|\e«6c oite a]a -peol/pojAit n-oeit^ 
toije &n T)eifV5 ueogonAro Dtnnt) ! 

— ubi vides omnia servari quae dicta sunt requiri : primó 
septem syllabis quodlibet constare quartum, dari concordiam 
in primo quarto primi semimetri, nempé inter acáIoi-ó et 
ot^Af. Item AtAioit) perfectam habere correspondentiam 
cum n--átic|\Átoit) ; sed et otjtAf et uocca correspondere inter 
se esto non tam perfecté, praeterea optime concordant zocza 
et U111L5 in secundo quarto primi semimetri, ubi ne superfluat 
syllaba eliditur a in cocca per subsequens a in av\. Item 
bene uniuntur cuitg et utunb quantum ad vocales, 
syllabas, sonum, et quantitatem ; vocula autem c^veuoc 
prima nimirum secundi semimetri et est et sortitur nomen 
unLAnn de quo supra. 

4. PoiTo 01 te et freotfogAit concordant, 01 te veró et toige 
bene uniuntur, sicuti et n-oeitg et -oeipj inter se ; similiter 
egregié uniuntur peoLfojAiL et tieojonAró. Ubi vides 
undique perfectum metri magni artificium, in se sane diffi- 
cillimi, ut et videre est in sequenti : 

CaVdac rr\AC Cacaoi|\ ná ccac, 
Za\\X)Aó 5^n ZAtAoip t>á tioj — 
"piAt av\ c-05, coriitttom ó ceAnc, 
^A]\ zzeAóz Congo it 'n-d Clot ! 

De Prosodiá Hibernicá. 55 

3. Let this be an example : 

"OeAtg At Atom otfAf "CoAxy^, 

X)Á1(\ n-AnCfAt010 coóca An ctnlg, 
Cfeuóc die aja peoLfogAiL nt)eil5 
loije An t)eif\5 Deojon.Ait) 0111 fb ! 
[Literally : 

111 comes upon ill — a thorn tears open Theigue's wound, 
At untimely hours come blows of ill fortune — 
After the havoc of the flesh tearing thorns 
Comes the payment of the bungling ignorant quack !] 

— Here you see all things observed which I said were 
necessary, each line consisting of seven syllables, an alliteration 
in the first line of the first couplet, namely, between AcÁLoró and 
otfiAf . Again we find AtÁlom making a perfect assonance 
with n-AnunÁtoib, there is also a correspondence between 
ot^Ay and cocca, though not so perfect ; moreover, cocc and 
0UI5 excellently alliterate in the second line, where lest 
there should be a superfluous syllable the a of cocca is elided 
before the succeeding a of at*. Also cuitg and buifb are 
well matched (or rimed) as to vowels, syllables, sound and 
quantity ; whilst the word cneucu, the first that is in the 
second couplet both is the u ft Ann (or shaft) of that line and 
receives that very name — concerning which see above. 

4. Further, oHe and -peol-pogAit accord (or alliterate) — as 
the aspirated p beginning the second word does not count 
either in pronunciation or prosody ; 01 te and ioi§e also are 
well matched, so also n-oeitj; and x>eM^s to each other : in 
like manner £eolfogAil and beogonAi-o are excellently 
matched. You see here the perfect art of the great versi- 
fication — in itself truly most difficult, as may be seen further 
in the following : 


Caaoac gAn CAtAOin *oa tioj — 
pAl A11 c-05, corhtjtom ceAnc, 
lAn cceAcc ó óonjott 'nA ciot ! 

Lit. : Calvach, son of Cahir of the battles — Serviceable beyond 
reproach to his house — Generous the youth, just from love 
of right — Right that falls in a shower from his sense of 

56 De Prosodid Hibernicd. 

Vel sic : 
*Oo cLeAcc mo 6a\\a 5Á cun 

ÚeAóc 50 z&riAX) teAcc 50 out) — 
tTlÁr a\\ st 1 ^' é 5 1on S^nt)' eA>0 > 
tlÁp An peA^ Atié Aniu'ó ! 

5. Quantum attinet ad aliud genus metri vulgó ^Annoi- 
geAcc oeAj, observa bene quod eodem plane modo fiat sicut 
prsecedens jAAnnoigeAóc rhóp, prseterquam quod omnia quarta 
in -jvAnnoigeACc ueAg finiant in voculam bissyllabam ut videre 
est in sequenti : 

ftojA ha cLomne Con Ait, 
C05A nA •opowse T>eA\\Am, 
U0L5 t>á|\ feotAt) j\tJ5 fotfiAm, 
Con ALL €U5 Ti'CojAn peA^Ann ! 

— in quo vides omnia observari quae observantur in praecedente 
metro, et nullam intervenire differentiam inter utrumque, 
prseterquam quod finales voculse hujus metri sint bissyllabae — 
istius veró finales omnes sint monosyllabae. 
Unde ad hoc genus metri attinebit sequens : 

On mAc j\ob AOfOA "o'piACArá 

UAOfgtA Ó5 mAt 5AÓ mtiA-oAij, 
CaLoac 05 -p°5 A Ari \úo%\\Art, 

UOJA CÍO|\tA1j CLÓ CUAnAlt) ! 

De Prosodiá Hibernicá. 57 

Or this : 
T)o cteACc mo caj\a 5Á cnn 

ÚeAóc 50 CAn<vo teAcc 50 unit) — 
íTlÁr An gnAt) é 5 1on S^nb' e^T), 
TlÁn An reAn Ané Anmt) ! 
[Literally : 

A friend of mine on invitation 

Used to act on the motto — " Come alone, come for long " 

But whether he meant it for friendship or not — 

I'd rather the guest of yesterday stayed not to-day !] 

5. Now as regards the other kind of metre called nAnnoi- 
$eAóc fce-Aj;, observe well that it is formed in almost the very 
same manner as the preceding nAnnoiseAóu mójt, except that 

all the lines in it finish in a dissyllable, as may be seen in 
the following : 

H05A nd clomne ConAUl, 

CojA nA "onomse "oeAnAtn, 

Uotj; T)Án reolAt) 1U15 norhAtn 

Con a VI ctig -o'eojAn -peAf Ann ! 


[Literally : 

Choice of the clan is Conall — 

Best of the tribe I declare — 

A chief for our guidance who went forward — 

Conall who gave Eoghan his kingdom !] 

— in which you see all things observed which are found in 
the preceding metre, and no difference occurring between 
them, except that the final words of the lines -in this metre 
are dissyllables* — whereas the finals in the former case are 
monosyllables. The following also will therefore be of this 
metre : 

On mAC nob AOfOA *o , '|piACAi > o, -]c. 
[Literally : 

From Fiacha's eldest son 

Little the good that falls on each worthy, 

But young Calvach is the choice of all chieftains — 

Best of patriots, with the true stamp of his race !] 

* And ending in dissyllables — the longer or greater points — it must 
be the true -pAnnuijeAcc rhófi, and not the other, which should obviously 
be called the -panmnjeAcc t>e&s. 

68 De Prosodid Hibernicd. 

6. Genus denique metri vulgo cAr-OAij\n in quolibet sui 
quarto septem syllabarum nnire debet in voculam trisyllabam, 
requiritque sicut dictum est de reliquis, suas concordias, 
correspondentias, et uniones proportione servata, nee minoris 
est artificii esto minoris apparent suavitatis, ut videre est 
in sequenti : 

pm^c fiío£ aoai* £ionnto$A, 

SÍ 0*0 CaÚAIL A CCOmlAttA, 

X)'& join t/Afvtn í lugAine — 
T)o rhA^u fom An fio*óui$e. 

— ubi vides quarta omnia septem syllabarum nnire in dic- 
tionem trisyllabam, concordare autem in primo quarto acato 
et ponntogA utpote incipientia a vocalibus, elisa litera -p. 
Item jvíog in ipso et -pío v ó in sequenti correspondent similiter 
AÓA1T) et CacaiL ; item CauaiL et ceo rh La* a concordant in 
litera c initiativa utriusque ; similiter in secundo semi- 
metro concordant n'^^m et UijAine. Item correspondent 
•o'a^ m in primo ejus quarto et mAfo in sequenti ; similiter 
50m et fom. Item concordant in postremo potn et po'dtiige 
in initiativa litera nempe consona f, etc., ut dictum est de 
Metro magno observata proportione, juxta hoc examina 
sequens ejusdem nempé metri carmen : 

CA0|\ A n^ADA 5fV1AntAf a ! 

A nAtiiA ó pei"óm AoiDneAfA, 
ZAmtA mvif fém piAf\AfA ! 

De Prosodiá Hibemicá. 59 

6. Lastly, the kind of metre called c-áfbAi|\n* — in which 
again each line has seven syllables — should finish each quartan 
with a word of three syllables, and requires, like the rest, 
its alliterations, assonances, and unions in due proportion ; 
nor is it less artistic, though it may appear less agreeable, 
as may be seen in the following : 

ptupc fúog ACA1T) pormtojA, etc. 
Literally : 

Seats of kings were the fields of Fionnlugh, 

Cahal's hill stood near their gates— 

But struck with the weapon of lughaine's grandson, 

That weapon slew the keeper of the hill. 
— where you see the lines, all of them seven-syllabled, finish 
in a word of three syllables, alliteration occurring in the 
first quartan between acai*ó and £ionnto&A, as they begin 
with vowels, for the -p is elided. So piog in the first line 
rimes with -pio*o in the next, and similarly acai-o and CaúaiL. 
CdtAit and ccomt^OA alliterate, both having c for their 
initial ; in the same way in the second couplet Aptn and 
1t)$Aine accord in having initial vowels. Also there is 
assonance between •o , A|\m in the first line of the second 
couplet, and rh^ftf) in the last ; and there is another rime 
between 50-m and -pom. In the last line foin and fío'ótnge 
alliterate in their initial consonant p : thus of other things, each 
in its place, as has been said of great versification. In addition, 
examine the following verse written in the same metre : 


Calvach 05 — great the warrior — 
On field of danger — falls like thunderbolt — 
Ever from hope of happiness 
Sternly he cuts off his enemy ! 

* CAffoAiftn : This being but a late form of the word, it would be difficult 
to explain its meaning if we had not older forms. It is a shorter form of 
cAft>Aifirie which was formerly cAf-bAi-jrone (i.)=ztwisted bardism or versi- 
fication, as a contrast, no doubt, to -oÁn -oíjieAc or direct metre. The same 
notion gave the name *oy&\ jneAc (thorny) to a metre with lines of nine or 
ten syllables, each ending in a trisyllable. From the short explanation 
given in the text, no very clear idea could be formed of its nature, except 
that the lines are seven-syllabled, all ending in trisyllables. Nor from the 
one example is it clear where the final rimes occur — apparently each 
couplet rimes. O' Donovan (Ir. Gram.) quotes the same example without 
translation or remark, but from other sources we learn that the final rimes 
were alternate — mostly the second and fourth lines. Another variety of 
CAf bÁijrorie was the otl-cAf bÁi-frone or cAfb. ceArmr jtorn (the great — or the 
heavy-ending cAf bAijvone) which had lines of eight syllables, each ending 
in a tetrasyllable ; but it does not appear to have been much used. 

60 De Prosodiá Hibernicd. 

CAP. VIII. [xxi.]. 
De Carmine vulgb ÓgtÁóAf. 

1. Jam sufficienter diximus de metro recto maxime nunc 
usitato, tametsi alia adhuc restant metrorum genera iis 
inferiora, et metra non recta ex quibus est prsesens. Metrum 
igitur, seu quasi metrum, seú umbram habens metri, seu 
metrum secundum quid dictum ó^tÁóAf fieri potest ad 
imitationem cujuscumque metri recti superius traditi, etsi 
frequentius fiat ad imitationem eorum quae vocantur T>eiDi*0e, 
feiron-d, metri magni, parvi, et non nunquam cAfo.Aij\n 
quantum ad primum semimetrum, quantum veró ad secundum 
ad similitudinem metri parvi. 

2. Omne ipsius quartum requirit septem syllabas, nunquam 
plures, nisi fiat ad imitationem -peu^tiA, tunc autem priora 
semimetrorum quarta constabunt octo syllabis. Simia enim 
est, et eó ornatius fit, si in singulis quartorum serve tur Con- 
cordia de qua superius dictum ; de vera autem unione, 
vera correspondentia nihil curat ; nihilominus, si inter- 
venerint, eo melius. Dixi de vera, quia similitudinarias 
seu apparentes admittit, atque requirit, hoc est ut sonet 
ad aurem, et loco verarum habeat Atrmf de quo supra. 
Exemplis res patebit. 

3. Accipe igitur óglÁóAf factum ad imitationem xieibroe 

ÍA "Óiuu 51L .dti ccttnn zú m $Áif 
'Sa coi$-fe Atnoij 50 tnó|Vó.áiL ? 
Hi binn tAif 5AÓ ctuAif t)o ctuin 
^amjk bAinfe ttnc m 'ÓÁturg ! 

De Prosodiá Hibernicá. 61 


Of the verse called CgtAoAf . 

1. We have already spoken sufficiently of *oÁn TrítteAó or 
direct verse — that now mostly in use — yet there remain some 
other kinds of metre, inferior to that and not direct, of which 
one is ósLÁCAf . This metre, or quasi-metre, or verse having 
a show of metre, can be made in imitation of any of the 
direct metres above described, although it is more frequently 
made in imitation of those called -oeioroe, yeux)r\&, tiAtintnjeACc 
great and small, and sometimes of odftt-Aifui, as far as the 
first couplet is concerned, but as to the second, in imitation 
of the minor versification. 

2. Every quartan of it requires seven syllables, never more, 
unless made as a sort of fetroriA, when the first line of each 
couplet requires eight syllables. For it is the ape of metres, 
and will be all the more ornate if each line has an alliteration, 
spoken of above, but of real union, or perfect assonance it 
takes little heed ; nevertheless, if they occur, it will be all 
the better. I say the perfect, because the imperfect or apparent 
assonances it admits and requires, that is, in order that they 
may appeal to the ear, and it may have Atrmf in place of the 
true assonances. This will become clear by examples. 

3. Take, then, this sample of ógtáóAf* made in imitation 
of "oeiDifte : 

A 'Óuid 51L aw cctuin cú m j;Áif\ ? etc. 
Friend Duff have you heard the shout 
From the house ith' field without ? 
Not everyone likes the ring 
Of young ó Daly's wedding ! 

* It is clear from the explanation and the example that ójtÁcAf 
is not really a different kind of metre, as -ftionnAfro, x>eit)i-oe, r-eu-onA — 
differing in syllables and rimes — but an inferior or inexact variety 
of any other metre, so that each of these has its ójtÁcAr- . The name 
is derived from 05IÁC, a servant, hence sometimes translated " service- 
metre " and less correctly "servile-metre." Originally, no doubt, it 
was the work of an ógtÁc, a young hand, an apprentice, a learner, 
not that of an oUatti or master. 

62 De Prosodiá Hibernicá. 

Revocatis enim in mentem quae dedimus de *oeibi"óe, 
videbis omnia hie currere non in rei veritate, sed quoad 
apparentiam quod sufncit, imó sufncit quod constet majoribus 
et minoribus extremis, et interveniat in singulis quartorum 
concordia, seú sana seú fracta, et interveniat Amur-, ita ut 
majus extremum contineat minus, esto ipsum excedat plus- 
quam uná syllaba ut hic : 

t)o|tb a tfie-dtAti ^r» 5^C cjvÁig, 
tliAtt niAc 6acaó ttliujítieA'óÁiTi ! 

— ubi vides contra legem et regulas recti metri vulgó -oeibróe 
voculam cjiáij; monosyllabam et minus extremum contineri 
quidem finaliter in vocula rhvnjrheA-oAm, velut in majori 
extremo, verum excedi plusquam una syllaba ab ipsa, si- 
quidem altera vocum est monosyllaba, quod in "oeibroe est 
vitium, non veró in ógtÁóAf , oeibi > óe, ubi Amur- supplet vices 
verae convenientiae. 

4. Aliud ógtÁóAf imitatur feu-ona de quo supra, in quo 
Atnuf habet locum, et collum vulgo bfiA^AiT) de quo supra ; 
imo si ponatur Atnuf cum vocula duarum syllabarum, eo 
melius, sed et sine hoc fit hasc simia, ut : 

"g&X) a Site 'n-AgAi'ó ch'Aisrn'O 
1or»Af\, pAttoinj;, -pitéx» f|vóit ! 

De Prosodiá Hibemicá. 63 

Recalling to mind what we have said of "oeibróe you will 
see all things occur here, not in the exactitude of that metre, 
but in appearance, which suffices, for indeed it suffices that 
there should be the minor and major points, that an alliteration 
should occur in each of the lines, either the perfect or the 
imperfect, and that there should be an Atmif or internal 
rime, and that the major point should contain the minor — 
although the former may exceed the latter by more than one 
syllable, as here : 

t>o|\b a tfie.dt.An Afi 5-ac CjiÁfg 

HiaLL itiac 6acac ttl u 1 5 rh e a*0 Am — 

[Proudly his foot treads every shore — 

The foot of Niall, son of Eochaidh !]* 

Here you see, contrary to the laws and rules of direct metre 

and of "oeibi-óe, the word cjaáij, a monosyllable doubtless 

and the minor point — contained, so to speak, in the final 

rhuijrheA-óÁin as major point, but it is exceeded by more 

than one syllable since the former is to be counted and is a 

monosyllable ; and in -oeibróe it is a fault, but not in the 

imitation of -oeibifte, when Atrmf supplies the place of a true 


4. Another ógtÁóAf imitates feuwiA, of which we have 
spoken before, in which Atnuf has a place, and also bjUjA-ro 
(or the " neck") ; indeed, if Amu-p be made with a word of 
two syllables, so much the better ; but with or without 
this, it still remains imitative, as : 

5 ad a Óíte 'n-AJAlt) Ch'Al^niT) 

1onAj\, -pAtloin^, piteo f^óiL ! 

[Wear them, Sheela, though ill-suiting — 

Vest and mantle, satin braid !] 
[This is not pure r-evroriA, because (1) the concords a§ai*ó and 
Aignro in the first line are not the true ones — the words 
should begin with different vowels : (2) pAtloin^ and -015™* 
are not perfect assonances but imperfect. The couplet is 
therefore o^tAoAf of feuwiA, i.e., imitated or imperfect 
r-euT>nA. But in syllables and in the terminations it resembles 
true fewonA]. 

* Or, possibly, " Fierce his wave (beats) on every shore — The strong 
wave of Niall," etc., for t-peACAn has at least two meanings, which 
probably point to two different words : (1) c-peAtAn, the foot, allied 
to the Irish ^013, cfiÁcc, Welsh troed, whence apparently the English 
tread ; and (2) cfieAÚAn, the sea, a sea-wave, allied to the Greek Triton, 
a sea-god. In the latter sense cfUAt is the usual form, cfieACAn being 
the genitive : it is to be noticed that at any rate the word above has 
a short eA, for eAcAc is the uAitne or internal rime corresponding. 

64 De Prosodiá Hibernicá. 

5. Ut fiat ad imitationem metri magni, nihil requiritur 
nisi quod extrema verba quartorum sint monosyllaba, ut 
ibi dictum est ; nee curat an uniantur fracté et imperfecté, 
anne autem perfecte et integré : sufneit prseterea intervenire 
Amur loco correspondent, verum postulat ut quarta indis- 
pensabiliter finiantur voculis monosyllabis, haec autem vox 
unius syllabi in sequenti quarto, cum qua habeat Amur seu 
apparentem correspondentiam ; et haec jaceat circa medium 
ejus, vel quasi medium et ut finales utriusque semimetri 
correspondeant saltern secundum Amur, ut videre est in 
sequenti : 

T)Á téi^tí 50 ruAimneAC -óó, ^ 

*ÓÁ *oi£ nó cj\í "o'ót T)Á cult), f ó^tÁóAr jvAnnuijeACCA 
*Oo oeu^AT) SeAAn *oo t)ÁL C " móij\e." 

A óeAX) *oo óác oeit Amu 15 ! ) 

Devotius autem sic : 

U|\i«f aca Ag bfiAit aj\ mo bAf \ 

510*6 acáit> T)o gnÁc im bun, ( óstÁCAr nAnn. 

UjuiAg 5AT1 a ccnocAT) f e cr-Ann f " móifve." 

An THAbAl,, An clAnn r & cnurn ! / 

6. Eodem prorsus modo fit óstÁóAr ad imitationem metri 
far vi, praeterquam quod dictiones finales debent esse bis- 
syllabae ut sic : 

A|\ *oo ctÁiffig 50 n*oúifie ^ 

tlí bí mo fúite acc *0|\ui€e, [ óstÁCAr f.Annuij;eACtA 

1 on Ann LeAm if a ctAirxnn C "bi^e." 

T)o tÁrhA X)'-pAiCfin uij\fie ! y 

De Prosodiá Hibernicá. 65 

5. In order that a verse may be in imitation of the greater 
versification nothing is required but that the ending words 
of the lines should be monosyllabic, as already stated. Nor 
does it matter whether they are united (rimed internally) 
in a broken and imperfect manner, or perfectly and fully : 
it suffices, however, if an Atmip occurs in place of a perfect 
rime, but it requires that the lines shall end in monosyllables, 
and the word in the second line with which the final of the 
first line is to have an Atntif or imperfect correspondence 
should also be a monosyllable ; and this should lie in the 
middle or near the middle ; and the final words of the two 
couplets should rime, at least according to amu-p, as may be 
seen in the following : 

T)Á téigtí 50 fUAitfmeAó t)ó, etc. 
[Give Johnny Wall to live in peace, 
And two or three pints of ale each day, 
Johnny Wall would leave the world 
Outside his cottage — there to stay !] 

or to take a more devout* theme : 

Ujvmf\ acá &§ tojvdit A^ mo uÁp, etc. 
Three who look for my death each day — 
Who'll rejoice when my day is done — 
(Would I could hang the three straightway !) 
The devil, the worm, and my son ! 

6. Just in the same way ó^tAóAf is made to imitate the 
lesser versification, observing only that the final words are 
now to be dissyllables, as here : 

A|\ "oo ctáifvr-i§ 50 troúi^e, etc. 
[On thy harp intensely gazing, 
Heart upraising now I listen — 
Thy deft touch makes matchless music 
Soothed my ears, my old eyes glisten !] 

* The wish expressed in the second couplet can hardly be called 
" devout," but Fr. ó Molloy was thinking of the general nature of 
the poem of which the above is the first stanza — a poem on death, 
of a moral or didactic nature, but with a strongly marked satiric vein. 
It is ascribed to T)ormcA-ó mójt oTDAIaij, Abbot of Boyle in the XHIth 
century, and is given in full at the end of this book. 

66 De Prosodid Hibernicd. 

7. Omne ójtió^f cujus primum quartum fit instar cAftiAij\n 
et secundum instar metri fiarvi, non requirit nisi ut voculae 
trisyllabae primi et tertii quarti finales habeant inter se 
convenientiam vulgó ^mnf : item ut dictiones finales primi 
et secundi semimetri talem adhuc habeant convenientiam 
inter se, ut videre est in sequenti : 

SlÁn «Aim "oon *oA Ao'óAifve, 

^Á Dptut An eA|\|A tiA "ocofAó — 

TVuArílAtl £1J\ ttA fAODtUICfl, 

Hi t)iu ni Af -piA t>á no*o.<vó ! 

De Prosodiá Hibernicd. 67 

7. Every o^Lac-Af verse whose first line is made like cAf- 
X)A\^x\, but whose second is made like the lesser versification, 
requires only that the trisyllabic final words of the first and 
third lines should have an Atnuf or imperfect rime, and that 
the p&dntiA or final words of each couplet should have the 
same kind of rime or agreement, as may be seen in the 
following : 

SlÁn tiAitn T>on t>Á ao*óa\^q, 

'^á opuit An eAjAj\ ha x>JzoyAó — 

X> y UAXf]AV\ flj\ tt-d fAODtlHCfl 

11Í diu ní av fiA t>á no'OA'ó I* 
[I salute two noble pastors, 
Shepherds true of flocks all faithful — 
Further noting I refrain from, 
For the naming might be baleful !] 

* Seems a cryptic message or salutation from the author at Rome 
to friends in Ireland — perhaps two of the Irish bishops — or perhaps 
the heads of two houses of his Order — whose more particular naming 
would be fraught with danger to those two friends. 

The verse above is mixed cAfbAifin and fiAnnuijeAcc " beA^ " : 
it is Ó3lÁcAf or inexact metre, because (1) there is no alliteration in 
the second line ; (2) there is no Amur- in the second or fourth line, 
and (3) the rimes of the first and third lines, and of the second and 
fourth, are the imperfect kind. 

68 De Prosodiá Hibernicd. 

CAP. IX. [xxii.]. 
De Carmine vidgb *o^oi5neAC. 

1. Hoc genus carminis vulgo -ofo-i^neAC, Latiné spinosum 
admittit ad libitum authoris in omni suo quarto indifferenter, 
vel novem syllabas, vel plures ad tredecim. Singula autem 
quartorum debent finire cum vocula trissyllaba. Finalis 
prseterea vocula primi quarti debet habere corhAjvoA de quo 
supra, cum alia vocula in initio, vel medio sequentis quarti, 
cujuscumque sit semimetri. Item uAicne suo modo sub fine 
secundi quarti ; reliqua autem verba debent inter se habere 
cotfiAjvoA de quo supra et convenire hincinde utrobique. 
Denique dictiones finales utriusque semimetri debent 
correspondere ut in sequenti : 

X)o jeit) -pom 3<in folXA jati imfveAjMin, 

^n rtój -pe pnnleAfAit) co-pcfA in óui^mtifin, 
'Sou "oíot one "oon uAtlCAt 6ATfmA-roin 
T)eAtt>tAfAif uuA'óctAc ttteToe nA t>tunjne-rin ! 

2. Nolo omittere aliud carminis genus, vulgo cAr£>Aif\n 
ceAnnqiotn, nimirum ex gravitate seu magnitudine capitis 
ita dictum : fit autem eodem prorsus modo quo ipsum cA-p- 
DAij\n de quo supra, prseter quam quod singula ipsius quarta 
desinant in dictionem quatuor syllabarum et quodlibet octo 
constet syllabis, ut videre est in sequenti : 

TTIac rú-o Af fViocc tionn-fTlAnAnnÁin, 
A5 fú*o An rtiocc f eAn^rhA^pAttÁin — 
A eAflA "ótfht TinornjLAn-poUtAm, 
Cúió Le n"oeAjwA *oeAt;-ttlAnAnn,5n ? 

De Prosodiá Hibernicd. 69 


Of the Metre called -onoijneAC. 

1. This kind of metre called -oyvoigne.Ao* — in Latin spinosum 
or the thorny — permits at the pleasure of the author any 
number of syllables in each line, from nine to thirteen. But 
each line should end with a word of three syllables. And 
besides this, the final word of the first line of each couplet 
should make a corhAjroA or rime with some word in the 
beginning or middle of the next line. There should be 
uAicne, too, or assonance of its own kind towards the end of 
each second quartan ; and the other words also should agree 
and assonate with each other, one in each line. Lastly, 
the final words of each couplet should rime, as in the following : 

"Oo $eit) nóm 5-Att folZA 5 An itnr\ eAr\4in, etc. 

[Literally : 

I find before me without quarrel or contention 

The family that owns the fair grounds of that festive 

court — 
Where the mere flash of the jewelled cups in the hall 
Will be sufficient light and fire to the proud race of 

Eamhain !] 

2. I will not omit another kind of verse — called car-DAinti 
oeAtincfiom, or heavy-ended cAr-DAinn, and no wonder, from 
the length or weight of the ce^nn or end of each line ; it is 
made, however, in the very same way as the CAfbAi^n 
already described, except that each line in it now ends in a 
word of four syllables, the whole line containing eight, as 
may be seen here : 

tTlAc rú-o Ar\ f tiocc "fMonn-ltlAnATinÁitt, etc. 
[Literally : 

Yon youth is of the race of fair Manannán — 
That was the race active and vigorous — 
But — O stout chief of the fine broad shoulders, 
By whom was made fair Manannán himself ? 

* This "oÁn •oji015rie.dc was, no doubt, called in contrast to t)Án -oíjieAc, 
the latter the direct or straight metre, the former the thorny metre. 
The word is derived from -orioijeAnn or -ojiai jeArm, Welsh draen, 
Eng. thorn, and it was probably called thorny from the irregular 
length of the lines, and from its alternate rimes — though, no doubt, also 
it was a difficult or thorny metre to compose in. I have seen occasional 
verses of ■ortoijneAc, but I am not aware of any lengthy poems composed 
in it — though such may still exist. 

70 De Prosodid Hibernicd. 

3. Aliud adhuc vulgó feuTmA rhórv fit quemadmodum 
ipsum retronA de quo supra, praeterquam quod semimetrum 
quodcunque hujus desinat in voculam trissyllabam ut : 

TVpor* co^Ait) corh.Ai1xe.Afi fíotcÁin, 
SeAnrocAt tiac rÁtnnjxeAn — 

in t«ftd fit A6c V e% rogtA ( r emM ™ *- 

peAt) t)Ánt)A tiA mbAnpoitrveA-o. 

4. Aliud adhuc restat vulgó reu-oriA rheA-oonAC, íitque 
instar seudnca praeterquam quod prima quarta utriusque 
semimetri finiant cum vocula trissyllaba, vel quasi tris- 
syllaba inter se concordantibus : postremse autem dictiones 
bissyllabse utriusque semimetri correspondeant, interveniat 
in quartis concordia imo et correspondentia quaedam inter 
finalem voculam primi quarti secundi semimetri et aliam 
sub medio quarti subsequentis : ut, 

peA-prx fitteA'ó da fAtm* neAtfrtHifóé ^ 

*Oo nítí ah teAbtoit) Litine — I feuTmA 

ÍTÍAifvj T)o seiD An jlóifv n-éACCArvUAi-ó J ttieAT)ónAó. 
Óit) Ar\ D|\éA5-f Atmoiu* binne ! ) 

5. Aliud vulgó fionnArvT) constat quatuor quartis, et omne 
quartum sex syllabi s, cuj usque finalis dictis est bissyllaba, 
ultimae metrorum correspondent, ultimum cuj usque quarti 
concordat cum aliquo vocabulo mox antecedenti; in ultimo 
praeterea semimetro debet intervenire correspondentia, ut 
in sequenti : 

Horn riA péite 'pÁriA-ro 
JpAirvCe pieAT) ei-peArm, 
5rvi^n tiA mAj An mionponn 
AnnArh 51 Alt ^Ati j;éioeArm ! 

* Ed. Rom. ^pfAlm," " pfAtmoib.'" 

De Prosodid Hibernicd. 71 

3. Another also, called p euwiA triors is made like p etronA 
itself, except that each semi-metre of this kind ends in a 
word of three syllables : as 

T)'pio|\ óo5Ai"ó corhAilceAr\ fíotóÁin, 
SeAnpocAt tiAó fArautjteArv, etc. 

[Literally : 

'Tis the man of war fosters peace — 

Proverb never belied— 

None finds peace but the well-armed 

At least in Banbha — though fair her woods !] 

4. Another still remains — that which is called p etroriA 
rheA-óónAó, the medium or intermediate petranA, and is com- 
posed like the simple petronA except that the first line of 
each couplet is to end in a trisyllable or a quasi-trisyllable, 
each of these to have the usual alliterations : the last words 
of each couplet, however, to be dissyllables riming with each 
other, with alliterations in each of those lines, and a certain 
assonance between the last word of the first line of the second 
couplet and another towards the middle of the following 
line : as 

peÁrvp fitteAt) ha fAttn neArivótiróe, etc. 

[Literally : 

Best for us to repeat the heavenly psalms 
That we used to hear by our bedside — 
Alas for him that prefers profitless earthly glory 
To the singing of the beautiful psalms !] 

5. Another called rviormArvt) consists of (stanzas of) four 
lines, each of six syllables, the last word of each line being 
a dissyllable : the last words of each couplet rime (that is, 
the last of the second line with that of the fourth) ; the pirm 
or point of each line alliterates with another word near it 
in the same line, whilst in the last couplet there should be 
besides a riming of one word (at least) in the third line with 
another in the fourth, at the beginning or in the middle : as 

Horn riA péite pÁnAit), etc. 

[Literally : 

Rome of the pilgrims is Fanat — 
Home of Erin's poets — 
Sun that lights fair regions, 
Land of conquering heroes !] 

72 De Prosodiá Hibernicd. 

Alii hoc juormAfvo vocant cj\ionn.Atvo ', instar ejus datur 
•aAtiAfVo* pauciorum adhuc syllabarum, plurium veró cic- 
TvennA|\T) ; et aliud adhuc cujus semimetra constant decern 
syllabis, quinque in uno quarto et quinque in alio — ut dictum 
est de Aodo ni fallor m^c th-oip, ut : 

^aw Aroroe x>'Ao*b 

1p amine An peuf ! 
Hoc est, non obediendo Aodo eo altius gramen ; in quibus 
vides quam eleganter servetur proprius syllabarum numerus 
et quartorum, necnon concordia, correspondentia et unio 
de quibus supra. 

6. Dantur et plura adhuc genera metrorum, vel rectorum 
vel quasi rectorum, de quibus quia raris et non adeo in usu, 
et mihi non occurrentibus, nequaquam loquar. His autem 
visis facile dignoscentur, et ipsa et eorum leges, adeoque te 
ad authores remitto, et ad veterum observantias librorum. 

*Ed. Rom. "T>tiAn.4jvo." 

De Prosodiá Hibemicá. 73 

Some call this not -piorm-dtvo but cf\iorm.dfvo* ; another 
like it, but of fewer syllables, is called T>&n&\vo, whilst one of 
more syllables is called cicj\erm,A|ro i there is also one of 
which each couplet consists of ten syllables, five in one line 
and five in another, as in the following, which was said, if 
I mistake not, of Ao-ó tn^c tlvoitt — 

IpAmme An -peiijA ! 
[i.e., Taller grows the grass 
Not regarding Hugh !] 

— which means " not having to obey Hugh, the grass is all 
the taller": in which you see how elegantly is preserved 
the due number of syllables in the lines, besides the alliteration, 
assonance, and union already spoken of. 

6. Other kinds of metre still are sometimes given — varieties 
of the direct metres or imitations of them, but as they are 
rare and not now in use, and as I have not met them, I will 
not speak of them here. From those shown here, however, 
others will be easily recognised, both the metres themselves 
and their laws, and therefore for these I refer you, O reader, 
to the authors and to the precepts of the old books. 

* This brief description of juontiAfvo is very far from satisfactory — for, such 
a well-defined, ancient and important metre should have had a chapter to itself 
instead of coming in as an afterthought in a chapter on "o-jioijneAc with which 
it has no relationship at all. The definition given, however, is correct so far 
as it goes, and the example chosen is a good one. The worthy father does not 
seem to have known the meaning of the term juonnAfeo, nor the meaning of 
•oAnAjro or cir-fiennAft-o, though he seems to recognise vaguely that the two 
latter had some reference to numbers, i.e., to the number of syllables— which 
is not the case. We need not wonder greatly at the author (who lived out of 
Ireland so long, and so faraway from books and documents) not fully realising 
the meaning of these old technical terms, when some of the greatest (at least 
most pretentious) scholars of recent times are still in doubt as to the meaning 
of some of them. It is certain Whitley Stokes in 1862, when he published 
the Glossary to the Feilire (Three Irish Glossaries), did not know what an 
Afro was. That, of course, is a long time ago, but eighteen years later when 
he published his first edition of the Feilire Aonghusa (1880) he was still in 
doubt, for he asks if an Áfra was not an alliteration ! The late Dr. Atkinson, 
who attempted to correct Stokes (and did on some points), made some 
blunders of his own on the same subject. 

The above is an example of fiionnA-|ro tftí n-Áfvo — that of the three rimings / 
these being (1) between éifieAnn and jéibeAnn, (2) between 5|UAn ua and 
31 Alt 3AU, and (3) between mAJ ah and AnnAtri. The metre in which the 
■peilifie Aonjxif a was written was not so elaborate as this — especially in the 
second couplet. [Translator.] 

I believe an Ájfo was a riming — not a particular word but the riming or 
agreement of two (or more) particular words. It is the oldest word, found for 
what was afterwards called comAffOAT) (later comAffOA) — cotriAjfOAX) itself 

74 De Prosodiá Hibernicd. 

CAP. X. [xxiii.]. 
De Carmine Hibemis t)f\tii tinge-doc. 

1. Hoc genus carminis fit cum correspondentia ad minus 
fracta seu similitudinaria de qua supra : item cum concordia 
aliqua et unione, necnon cum extremis et capitibus ut supra, 
suo modo. Quarta rursus potiuntur septem syllabis ; ipsum 
fit ad imitationem cujuscunque metri recti, sed ex omnibus 
imitari solet metrum vulgó cdr-udifin nuncupatum, ut : 

triii c CA0U115 a% ctdr-tnjedcc 
P-d uun dotctnn c'edf ccajvai> : 
Verum mihi videtur quod imitetur potius carmen vulgó 
fetrona rhe.d'óónAC de quo supra ut ibi videre est. Hoc alio 
autem modo imitari potest metrum parvum, de quo supra : 

AcA a tAn 'o'ioó'o if ^'emedc 
1fm Cfvior acá toy^AX) ! 

2. Eodem autem modo fit utrumque semimetrum ipsius 
quando fit ad imitationem metri parvi vel metri cd-podif\n 
et cognominatur ab ipsis generibus quae imitatur, sicut et 
cognominatur ógtÁóAf ; continebitque tot et tales syllabas 
in quartis, quot et quales requirit metrum quod imitatur, 
servatis hinc inde concordiis suo modo, correspondentiis, 
et unionibus, necnon convenientia utriusque semimetri, ut : 

tTlÁf 'oeom -piiAftif nó éigedn 
A ^íogAin •óéi'oje.dt *úat:a, 
X\r coi|\ mo ceng-dt x^'a^ax) 
tló mo mAfX)At> te X)aza ! 

being but a derivative of Áffo and equivalent to co-riming or cor-respondence. 
rtionnAtro was |tin-o-Án"o=point-riming, i.e., the riming of the dissyllabic 
jie-AnnA of the second and fourth lines — one of the characteristics of that 
particular metre. I believe we have in our Irish Ájro (whether anyone else 
has pointed this out before I don't know) the exact analogue of the Latin art- 
(nom. ars). It is also feminine in gender like the Latin ays : gen. Áijroe, 
d. Áiffo, n. plu. djvoA ; so ftionnAjro, fiionnAiffoe. The root of the Latin 
ars, arma, artus, etc., did not exist in Latin, but it existed in Greek in the 
reduplicate áp-ap-/6X0J, future áp-w J and the root of Ájro, A-pm and other 
allied words existed in the O. Ir. Aft-im, Aifi-im, now oifi-im=I fit, suit, agree, 
oi|ieAtriAin=agreement, oijieAmnAc, fitting, suitable. In fa\vo-pir)n==majUS 
extremum, we have on the contrary the ordinary adjective Á|fo=high, long, 
a different word altogether and allied to the Latin ard-uus. A third Áfro or 
Áfic (now Áfro or Áifvo f.) means part, point, direction, and is allied to the 
Latin pars, part — with loss of the initial p. CoriiAfroAX) or rime was one of 
the oldest characteristics of Irish verse, and in the use of the old word Afro 
for riming we have a special kind of art — namely, the harmony or agreement 
x){ words and verses. Compare "oÁn a poem, originally any art, or artistic 
work ; and rime in Milton's line "To build the lofty rime," i.e., the lofty poem. 

De Prosodid Hibernicá. 75 


Of the verse called by the Irish "DjuntmseAcc. 

1. This kind of verse is made with assonance, at least of 
the broken or imperfect kind spoken of above, also with 
some alliteration and union, besides the points and endings 
already described. The lines as before have seven syllables 
each, and the metre is made in imitation of each of the direct 
metres ; but of them all, the most frequently imitated is that 
called c-dfttAifm : as, 

TT!uc CA0U115 .d<5 cLAftnje-dec 

1pA otm Aotctii|\ z e-Af cca^ax) : 
[Like a wild hog trespassing 
Near thy foe's fair dwelling-place !] 

But to me it seems that this example imitates rather the 
verse called feuwiA rhe,<vóónAc, of which above, where it may 
be seen.* In another way, however, it may imitate the 
lesser versification (jvAnrmige-Aoc fce-dg), as follows : 

AcÁ a lÁn TMocT) if ^'eineAC 

1f1t1 Cf\10f &CÁ COfAT) ! 

[Within that vesture simple 
Beats heart of noble nature !] 

2. And in the same way is made each of its couplets when 
it imitates either the lesser versification or adfttatgnj and it 
is named respectively from the kinds of metre imitated, as 
ogtAc^f itself is named ; and it will contain as many syllables, 
and of the same kind, in each line, as the metre imitated 
requires, all things being observed, each in its own way — 
as the rimes, alliterations and unions, as well as the final 
assonance of each semi-metre : as 

TTlÁf Tíeoin puAjuif nó éige^n, etc. 

[Whether of thy free will, or by force, 
O fair queen of the pearly teeth — 
Thou may'st bind me in silver chains 
Or slay me with thy baton !] 

* The example, however, is not feti-oriA iiie<yóóiiAc but cArftAiftn — 
or rather its ó^lÁcAf or bfuiitin^eAot: — for fe-u"oriA meA-ó. has eight 
syllables and seven alternately, the eight-syllabled line ending in a 
trisyllable, and the seven-syllabled in a dissyllable. (See Chapter on 

76 De Prosodiá Hibernicá. 

Vel sic : 

mtmA jaaid méin rtiAit T)Á cóin— f D r i11 Lin 5 eACC 

tlí fruit r an ccftot fe^onn fu^c ?- ^nnui 5 eAóCA 
Acc ctoi'óeArh tuAifte a ccfu-diu, óip ! J rooipe. 

Instar praeterea Metri magni componitur hoc modo : 

t)ío f FOfi f Um$ An oáta Mm ! ( ^uitinse^c 
tUó é ^r -oomín >oo 5 aC aoti f tumnuiseAóCA 

t)Ait a mbéiT»íf A AOf 5^ 1>0 ? / ™ 01 F e - 

3. Vel sic cum quinque syllabis in quolibet quartorum : 

"Uó -df ciaíi mo cuAijAC \ 

te t)Á u&ip T)o tó — f bpmlm^eACz 

that) mo piAn m-Af mó ! J 

4. Hoc similiter modo imitatur adfDAif\ri finiendo nimirum 
quarta in trissyllabis, servatis reliquis ut suprá : 

1£&t>a AzÁm tiA n-itijeAtiA 
*Oo Cu&vó Af óe-átin tiA cjuuce — 
tlí né rnoitte -áti imteACTM 
Ace a cceAgrhAit f a mtnne ! 

Sic proportione servata, ut superius jam insinuavimus 
aliorum potest imitari naturam ; de quibus non pono exempla, 
quia legenti et observanti hucusque dicta statim íient mani- 

De Prosodiá Hibernicá. 77 

Or take the following as an example of the bpuitinse-dcc of 
fv.Annu15e.ACC tíión : 

Hi fruit f aw $tó|\ tjpAoiti* x\-a\z, etc. 
[In the pleasant outward speaking 
Nought is there unless truth's told — 
Pleasant form and soul all sinful, 
Leaden blade in sheath of gold !] 

Imitation of the greater versification is also made in this 
manner : 

UpuAj An t)Áit, ón cjui,A5 aw *oÁit, etc. 
[Sad the fate, oh sad the fate, 
Sad man's state where'er he wends — 
True it is men have no home 
Where'er they roam, except with friends !] 

3. Or thus, with five syllables in each of the lines : 

U6 Af ciAn mo ctiAipx, etc. 

Long have I journey'd 
For two hours of day — 
Should I go further 
The dearer I'll pay ! 

4. In the same manner cAfDAipn* is imitated by finishing 
the quartans in trisyllables, other things being observed as 
above : 

1£at>a at:áyo wa h-inge-An-d, etc. 

[Literally : 

How long the girls are coming 
Who went to fetch my harp ! 
Kept not by length of the way 
But by their gabble in the garden !] 

Thus by keeping due proportion as we have already implied 
one can imitate the form of the others of which I do not give 
examples, because to him who has read and observed what 
I have said thus far, they will become at once manifest, f 

* This last example, however, is rather f eu-otiA theA-oonAc, with its alternate 
eight and seven syllables, than seven-syllabled CAfbAipn. See the beginning 
of this Chapter, par. 1. 

+ From this Chapter on bftui lingerer we gather that it is not a special 
metre, like fuontiAfvo or "oeibi"óe or peyonA — but that, like ójlÁcAp, it is 
rather a free and inexact imitation of other metres, and how it differs from 
Ó5IÁCAP is not very clear. The name bfiuilmjeAcc (like CAr-bAifin, 
CfUOrmAJVO, etc.), has probably become much changed from the original form : 
the author himself spells it bfUnlmseAcc and btuiijlingeAcc ; bjtulmseACC 
and b|\uiltineAcc are other forms. It is to be noticed that though the author 


78 De Prosodiá Hibernicá. 

CAP. XL [xxiv.]. 
De^quibusdam pracognitionibus. 

1. Diximus alibi aliquatenus de quibusdam necessariis et 
necessario praecognoscendis ad componenda hujusmodi car- 
mina, nimirum de divisione, de conjunctione, et amnitate, 
et mutabilitate et eclipsi et potestate consonarum ; item 
de earum elatione, de brevi longa et mediana quantitate 
vocalium, quae omnia sunt valde retinenda, sicut et divisio 
vocalium in largas et subtiles, de dipthongis et tripthongis*, 
hoc est bivocali et trivocali ejusdem syllabae. 

2. Diximus praeterea consonarum alias esse leves numero 
septem ; alias veró molles numero tres ; alias duras numero 
tres ; alias fortes numero quinque, tametsi hae quinque 
aliquando evadant leves appositione signi longi : alias asperas 
numero tres, aliam vero dici classem, etsi non mereatur tale 
nomen, eo quod contineat solam consonam p quae proinde 
vulgó appellatur solitaria et omnium regina, quia nobilissima 
et conjungibilis cum quacumque, et prae omnibus requirens 
in correspondentiis aliud f sibi respondere. 

3. Adverte diligenter secundum poetas ex his consonantibus 
alias aliis esse nobiliores, utpote potentiores ; post f enim 
molles dicuntur c, p, c, nobihores quam durae, quam fortes, 
quam asperae, et quam leves. Similiter consonae durae b, x>, 5, 
potentiores sunt asperis, fortibus, et levibus. 

Pariformiter asperae sunt fortiores levibus et fortibus, 
fortes veró fortiores sunt levibus, adeoque leves sunt omnium 
ignobilissimae et debilissimae. 

gave the name of this verse before ó^LÁcAf in Chap. I. he has treated the 
subject after it ; from this circumstance, from its being imitation like o^tACAp , 
and the name ójtÁc-df being derived from ójtÁó, a youth, young man, 
young soldier, I should not wonder (speaking without any knowledge of the 
ancient forms of the word) if it were formed originally from bfivnnneAt or 
bjimnnjeAt/, a maid — as if maiden's work, implying originally a less 
vigorous and less exact versification than even ójtÁcAf — if, indeed, the word 
is not a corruption of bAtrt-ptVfoeACC (woman's versification — the work of a 
bAin£ile or poetess). 

* Sic in Ed. Rom 

De Prosodiá Hibernicá. 79 


Of certain things to be known beforehand. 

1. We have spoken elsewhere at some length of certain 
things that are necessary and that must necessarily be known 
for composing in these metres — of the classification of con- 
sonants, of their combination, of their connection, of their 
mutability, of their eclipsing, and of their various degrees 
of strength ; also of the lengthening of vowels, of quantity 
short, long and medium — all which things must be well borne 
in mind ; as well as of the division of vowels into broad and 
slender, of the diphthongs and triphthongs — that is, of two 
vowels and of three vowels occurring in the same syllable. 

2. We have said, moreover, that some of the consonants — 
seven of them — are light ; that three others are soft; three 
others hard; that five are called strong (or heavy), though 
these sometimes become light when marked with the sign 
of length* ; three others are called rough ; and another 
spoken of, indeed, as a class, although not deserving the name, 
for it contains but one single consonant p, which for that 
reason is called the solitary and the queen of all the consonants 
as being the noblest, and capable of being joined to any 
other, and above all things requiring in its corresponding 
words another r to match it. 

3. Notice carefully that according to the poets some of 
these consonants are reckoned nobler than others, and so to 
say, more powerful ; for after r, the three soft ones c, p, c, 
are said to be nobler than the hard b, t», 5, than the strong 
(it, nrt, etc.), than the rough (c, p, t, f, -p) and than the light 
(X), "Ó, $, t, n, n, rh). Similarly the hard (o, t>, 5) are more 
powerful than the rough, the strong, and the light. So also 
the rough are stronger than either the light or the heavy, 
the heavy stronger than the light, and the light therefore 
are the meanest and weakest of all. 

* Consonants are not marked with any " long sign " ; the strong 
consonant tn becomes light when dotted or aspirated, but the other 
strong consonant-sounds (It, tin, fi-p, nj) take no sign when made light. 
At the beginning of words t, n, ji are naturally strong and are made 
light when other consonants would be aspirated, as after the possessive 
pronouns, certain prepositions, etc., as mo tÁm, mo neAfir, -jc. Haliday 
was, I think, the first who distinguished these sounds in printing : 
thus tÁm (=llÁm) but mo Uatti (with two points beside the light 
or "aspirated " t). 

80 De Prosodiá Hibernicá. 

4. S omnium princeps sibi non admittit parem neque 
similem, neque affinem neque cognitam, neque ullam ejusdem 
secum classis, tametsi cujuscumque consonae consortium in 
eadem syllaba, unde in metro cum nulla alia facit concordiam 
nisi cum alia p ut dictum est. 

5. Nota quod tn in voculis finiens raro constituat syllabam 
longam, nisi in paucissimis signatis pro longis ut mÁtn, AcÁim 
vel cÁtn ; imó consonae fortes in fine voculse ordinarié mediam 
retinent quantitatem in accentu mediam inquam inter brevem 
et longam, ut alias insinuavi. 

Quando n sequitur immediate ad p, tunc ph sonat eodem 
prorsus modo quo / apud Latinos, adeoque in metro non 
solum p cum p sed insuper pti cum p facit concordiam, ut 
videre est in sequenti quarto metri : 

A*orfiAim "ÓU1C mo fieACArt péin ! 

6. Quotiesumque íi sequitur ad p in initio dictionis inter 
quam et aliam debet intercedere concordia, nullatenus atten- 
ditur ad p vel ti, sed ad subsequentem literam, sive consona 
sit sive vocalis, ut talis dictio concordet cum altera incipiente 
ab aliqua vocali similiter, vel consona — ut alibi dictum est, 
ut videre est in sequenti quarto : 

U-A^Aip teAtn a pLAit éiptie ! 

— ubi propter t in teAtn et t in ptAit non obstante p, datur 
perfecta concordia, sicuti inter t in frtAit et t in tipe quarti 
sequentis : 

UA^Aip teAtn a ptAit tipe ! 
ut alias dixi. 

De Prosodiá Hibernicá. 81 

4. S, the leader of all the consonants, does not admit an 
equal nor even one like it — has neither kith nor kin, nor any- 
other letter with it in the same class, although it unites with 
any other consonant in the same syllable ; and therefore 
in verse it makes no correspondence except with another p , 
as said before. 

[Correspondence here implies both assonance (tear ue^f, 
piop, cpiop) and alliteration (as púit, pAL, -pion, peot)]. 

5. Notice that m final in words rarely ends a long syllable, 
unless in a very few which are marked as long as mAm, 
AcAitn or cAm ; nay, strong consonants in the end of words 
ordinarily give the vowel a medium quantity in pronunciation 
— medium, I say, that is between short and long, as I have 
already pointed out. 

When ft immediately follows p, then pft sounds exactly 
like the / of the Latins, and therefore in verse not only p but 
pft also makes an alliteration with p, as is to be seen in the 
following metrical line : 

xVotiiAitn *oui€ mo fieACArt pém ! 
[I confess my sins to thee !] 

6. Whenever ft follows p in the beginning of a word with 
which and some other there is to be an alliteration, the p 
is in nowise considered, nor the ft, but the subsequent letter 
whether consonant or vowel, as such word can only accord 
with another beginning with an aspirated p or with a vowel, 
as I have said before, and as may be seen in the following 
line : 

U^5Aip te-Am a flAit éiptie ! 

Where on account of the t in te^m and the t in ptAit there 
is a perfect* alliteration, in spite of the aspirated p : so also 
between t in ptdit and the t in bpe in the following line : 

UA^Aip leAtn a ft Ait tipe ! 

—as I have said elsewhere. 

* Rather the weak or imperfect alliteration — occurring, that is, in 
the earlier part of the line ; that in the second example being a perfect 
or strong one, as it occurs in the last part of the line. See Chapter III. 
for the distinction. 

82 De Prosodiá Hibernicá. ' 

7. Ad hujusmodi concordiam cum consona p requirítur 
in alia vocula similiter p initiativa indispensabiliter, nec rn 
(f ) concordat cum f , sed cum pti similiter incipiente ; nec 
r-L nisi cum rt, nec rn nisi cum r-n, nec pj\ nisi cum r-jv Si 
tamen casu contingat quod f initiativa voculae sequentis ad 
articulum An, ut aw crúit taceatur, tamen ut concordet, 
requirit in altera vocula r initiativam, alias vitiabitur metrum. 

Dantur quaedam consonae quae ad consortium alterius in 
syllaba mutant suam naturam et accentum ut mox dicemus. 

8. Alias do regulas pro correspondentiis : 

i. Ad correspondentiam non licet uti una consona simplici, 
nisi correspondeat ei aliqua saltern ex sua classe, sivé sit 
ipsa, sivé alia ejusdem classis, ita ut levis correspondeat 
cuicunque levi, aspera asperae, et similiter de aliis. 

ii. Ut detur correspondentia non requiritur ut fiat hinc 
inde inter plures consonas repertas immediate in una dictione 
quam inter duas cum alia dictione etiamsi ex ipsis una sit 
fortis et altera levis, adeóque bene et sufficienter correspondent 
pojrhAf et ^o^mjtAn. 

De Prosodiá Hibernicá. 83 

7. For an alliteration of this kind with the consonant f 
there is indispensably required another initial f in another 
word, nor does f alliterate with f but with another aspirated 
f , nor does ft accord with anything but ft, nor fn but with 
fn, nor f|\ but with ff*. If, however, it happens that an 
initial f is preceded by the article An, as aw cfuit, and is 
silent, then there is required in another word another f 
(similarly eclipsed), otherwise the metre will be violated. 

Some consonants when joined to others in syllables change 
their nature and pronunciation, as we will presently show. 

8. I give here some further rules for assonance : 

i. For assonance it is not allowed to use a simple consonant 
unless there corresponds to it another at least of its kind, 
either the same consonant or another of the same|class, so 
that a light may correspond to a light, a rough to^a rough, 
and so of others. 

ii. When an assonance is made, it is not necessary that it 
should be observed between more than two of the consonants 
found in the middle of a word, even if one of these should 
be strong, and the other light ; so that f o§riiAf and gofmistAn 
rime correctly and sufficiently, f 

* No doubt this was the best and most perfect alliteration where 
f was concerned, and probably was observed in the early ages, but 
in medieval Irish verse r alliterates not only with f pure but also 
with r followed by a liquid — ft, fn, rjt — but never with fc or fp, or 
re. Thus O'Huidhrin frequently has lines like — 

" CftiAtt uAf t)eAftt>A riA ff eAt> re^n " 
and " ó DeAftbA co StÁine roift ," etc. 

t Probably when the consonants 5111 in the middle of f. ojtiiAji were 
more strongly pronounced, this word was considered a sufficient rime 
to 50ft mijtATi : but for the last two hundred years these words would 
not be considered good rimes. For in 30t1m5t.An there are four internal 
consonants still pronounced everywhere — even the aspirated 3, though 
this not so strongly ; but in f 03™^ the 3 is generally silent altogether, 
whilst in Munster both 5 and m are mostly silent, and the word is 
pronounced pó-Aft, almost as a monosyllable — but^with a slight trace 
of the nasal m. 

84 De Prosodiá Hibernicá. 

iii. Non licet in altera correspondentium vocula poni p 
aliquam superfluam, seú cui non respondeat in altera adhuc 
alia f , unde vitium est in sequenti versu in quo non bene 
correspondent ccÁifgfin cum jvÁitfin. Spernatur ergo author 
qui cecinit male : 

tlíf toif5 *oon coif5fin x\a age — 
toifSpit) "oon coifs 01 te \at> — 
'OAoine X)Ax\ cCÁif5fiíi ^An cjwo 
Sati jiÁitfin cjvdoiDe C|\u^óon ! 

— ubi vitiatur correspondentia inter cCÁif^fin et yMtym 
vel certé inter coifsfin et Loirspró. 

iv. Ut fiat correspondentia consona mollis sivé sit c, sivé 
p, sivé c, admittit consortium cujuscunque consonae durae 
fortis vel levis in una dictione correspondentium ; et quae 
conveniunt cum alia consona molli, et altera quacumque 
posita in ipsius consortio praeter p ; adeóque bene corres- 
pondent feAócrhAin et teAncAi|\. Item ^e^n-cnúit et fe^n- 

v. Quando consona aspera, nempé c, vel t (vel f) vel p, 
vel f sequitur immediate ad levem, nempé ad t>, rh, "ó, 5, 
t, f, n, tunc sonabit instar consonae mollis qualis est c, vel p, 
vel c, ut videre est in -piAccott, quod sonat ac si scriberetur 


De Prosodiá Hibernicá. 85 

iii. It is not allowed in one of a riming pair of words that 
there should be a superfluous p , or one to which there is none 
corresponding in the following word ; whence there is a fault 
in the following verse in which cCÁif5r lri and f\Áitfin do not 
well assonate. Let the poet therefore be condemned who 
sang badly as follows : 

Tlif loif5 *oon coif5fin tia a§e, etc. 
[On that day he burned no houses — 
But he'll burn when next he'll war — 
Creatures without goods or cattle 
Shunning battle fled afar !] 

— Where true assonance between cCaifgfiti and jváitr-in is 
spoiled, and indeed that also between cotfSftn and toif^p-ó. 

iv. In order to make an assonance, a soft consonant c, p, 
or c, admits a union with any other, hard, strong, or light, 
in one of the riming words, and these are to agree with 
another soft consonant and some other joined to it, except 
p : so there is a right correspondence between f QAtzmAm and 
leAnu.Aifv, also between geAncnúic and pe-dticúift. 

v. When a rough consonant, namely c or t (or -p)* or p 
or f follows immediately a light one — u, -o, §, rh, t, r>, j\ — 
it will then sound as a soft consonant, i.e., like a c, c, p, 
or f respectively, as is to be seen in f iacóoVI, which sounds 
as if it were written fiAccolt. 

[Not a good example, for the aspirated c does not here 
follow a light but a soft consonant c, and pLac-coIL would 
not now be pronounced with a pure c, but pLAC-colt, and in 
Munster with a slight vowel sound between the syllables 
ftAC (a) Colt. Compare p e^n (a) -poc, pirin (e) -oeAn, An (a) 
-tfiAit, etc., a pronunciation which seems to point to the 
former existence of a connecting vowel between the 
syllables] . 

♦The aspirated p (ph or p} is certainly not a "rough" consonant 
like c, i, or p — in fact, it is the t pure p (being a strong aspirate) which 
is the " rough " consonant : the author himself has elsewhere shown 
that p in prosody is treated as p. The aspirated p was silent not 
only in Fr. ó Molloy's time but for many hundreds of years before — 
and the point over the p was really the punctum delens. In ancient 
Irish the p appears to have been pronounced like modern Irish t> or 
English v ; we find bAft and pAji, oaíIa and paILa, and in the future 
p is a development of b : thus mot-pAt) comes from O. Ir. mol-Ab or 

86 De Prosodid Hibernicá. 

vi. Consona dura nempé t>, t>, 5, si venerit ante asperam 
nempe cii, cti, vel p vim habebit consonse mollis, et aspera 
vim consona? levis, ut videre est in 5 ao Cop, b\\Az^At> [op&x>- 

CfA* ? ] 

vii. Ubicunque cti venerit immediate ante vel post f 
sonabit instar levis ut videre est in ce-Aó-fo, peA^-po quae 
et bene concordant ; similiter die de cne^f C-Aorh et cj\eAf Laoj. 

viii. S potestatem habet mollificandi consonas, seu reddendi 
molles positas ante se quinque, nempé t>, t, b, x>, % \ indurandi 
veró seú duras reddendi positas immediate post se, ut in 
pte-Af ctiof j\ ; enunciabitur enim ac si scriberetur pteAf-50^. 
Ex dictis sequitur quod bene concordent in metro teAcpo et 
pte-Aftfo. Item ce.Apr-0 cAT)fo quae tamen non correspon- 
derent bene nisi ratione potestatis quam habet litera f supra 
antecedentem, et subsequentem immediate consonam, non 
enim correspondent -pteA^ó et ceAp, et sic de aliis, quia p est 
mollis et *ó est levis. Litera autem f immutat eas ut dictum 
est, facitque concordare. 

De Prosodiá Hibernicá. 87 

vi. A hard consonant, namely t>, r>, or 5, if it comes before 
a rough (c, p, t or p) will have the force of a soft,* and a 
rough the force of a light one, as may be seen in 5.<vocop, 

UpACpAt) [bpAO-pAÓ ?]. 

vii. Whenever c comes immediately before or after p it 
will sound light (5) as may be seen in ce-dc-pof pe^p-po 
which make a good assonance, you may say the same of 
cneApóAorh and cpeAplAog. 

viii. S has the power of mollifying consonants placed 
before it >ó, t, t>, *o, 5 ; but of hardening those coming after 
it, as in pteAp-copp — for this will be pronounced as if written 
pteAp-^opp ; from which it follows that LeAcpo and pteA>opo 
make a sufficient agreement. Also ceAppo and c^vopo — 
which, nevertheless, would not rime well if it were not for 
the power the p has over an antecedent or subsequent letter : 
for pte^ó and ceAp do not rime, and so of the others — because 
p is a soft consonant and t> is light. But the letter p changes 
them, as has been said, and helps them to accord. 

* This is noticed in past participles and other verbal forms where 
an aspirated r (i.e., i) gives a preceding b, x>, or 5 the sound of p, z, 
or c respectively ; as fcuAb-tA=" fcuAp-A," 3oi-o-te="5oir-e," teA5-tA= 

"teAC-A," -]c. 

+ This can hardly be said to be the case now at any rate ; ceAc-f o 
may also be written reAtj-f o because teAj; is another well-known form 
of the word ceAc, but mo cpeAc-fA is certainly not pronounced " mo 
c-fieAj-r-A." CneAf-cAom is also unusual : after p it is customary for 
c, 5, c, "o to remain unaspirated, as eAf-cApA, -oeAf-^AbÁit, CAf-rA, 
*OeAf-mumA, -jc. 

88 De Prosodiá Hibemicd. 

ix. S fmalis cum alia p accedente ut in stAf-f a valet ad 
correspondendum, stante hujusmodi duplicitate alteri dictioni 
habenti unicam f ut Toa^a. Similiter consona dura et aspera 
correspondent quando molliuntur adeoque correspondent 
íoca et íocóA. Similiter duae consonae leves deserviunt ad 
correspondendum uni consonae levi, ut patet de uAi-óm et 
5u.Aitni£> in quorum primo invenitur t> quae est unica consona 
levis, in secundo autem reperiuntur in quae sunt duae consonae 
leves. Unde metrum : 

T)ot u.avoi£> ni but) "010^ t>Ari) 
Ap 5iíAitmb HÍ05 Ay jviojAn ! 

Similiter quando --da venit post aut t aut n, facit con- 
sonam fortem et correspondentiam cum forti, ut inter majus 
et minus extremum ; adeóque hinc minori extremo u^a 
bene concordat majus extremum r-Ao^ufoA. Similiter die 
de ipso respective ad alias leves. 

x. Quando duae vocales co-eunt simul in duabus syllabis, 
ita ut unam non constituant sed duas, prima erit brevis, si 
unquam fuit natura brevis. De syllaba autem una bivocali 
vel trivocali, quoad quantitatem attinet omnis bivocalis ao, 
ti a, eu, et iA, et omnis trivocalis aoi, imi, et nil, erunt longae 
quantitatis, vulgó fi$ne vax>a. Similiter longa erit Ae semper 
et ordinarié excipe voculas quibus adjungitur, ut aiunt 
Hiberni tanquam ctu^mf^ne, ut jonae et similibus in quibus 
fit brevis. 

De Prosodiá Hibernicá. 89 

ix. A final f with another in an added syllable, as in ^LAf-f a, 
avails — even with this double f — to rime with another word 
having but a single f , as b-df-A. So also a hard and a rough 
consonant when softened will correspond, as íoca and íocóa. 
Similarly two light consonants serve to rime with one light, 
as is clear from uArtm and 5114111110, in the former of which 
is found t) which is a single light consonant, while in the 
latter we have in two light consonants. Whence this couplet 
is correct : 

T)ot UAiftib ni but) *oíoja *óArh 
An 5UAitnib f\ioj &ip nioj^n ! 
[To leave ye thus would mean much wrong, 
To leave such throng of kings and queens !] 

In like manner when the affix -x>a comes after t or n it 
makes the consonant strong, and these letters (voa, viva) 
correspond with another strong consonant followed by a as 
in the greater and lesser point ; hence to the minor point 
u^a the word -[M0511V0.A will make a good major. This 
may be said, too, of the same ending after other light con- 

x. When two vowels meet together in two separate syllables 
so that they do not constitute one but two syllables, the first 
will be short, if ever it was short by nature. Of a syllable, 
however, with one diphthong or triphthong in it, as regards 
quantity all these diphthongs ao, ua, eu, and ^A, and all the 
triphthongs Ao^ ) uai, tut, will be long, without being marked 
by the fijne ipAX>A, or long accent. So also ^e is usually 
long, except in a few words in which it is joined as a personal 
or verbal ending — what the Irish call a cui*oinr5ne — as in 
goriAe, in which and similar words it is considered short.* 

* I have not met this, but it appears to be an old subjunctive 
form — the second singular of the present tense — for jone (cf. bejie, 
CAjie, etc.) and is probably due to teAtAti te LeAtAn — {i.e., joriAe for 
jone). So the ending of the past participle -ca, -ca is sometimes 
written -rAi, and -tAe, both of which were probably short (cArcAe, 
beftcAe, jutAe, etc.). 

90 De Prosodiá Hibernicd. 

xi. Bivocalis eo ordinarié semper longa est, exceptis paucis, 
ex quibus est *oeoc, neoc. Bivocalis qa semper vel brevis 
est vel media, praeterquam in cuiTrnirgne secundae personam 
ut cum dico x>Á mbu-AitteA ubi longa est. Caeterúm naturá 
brevis est si non reperiatur inter bivocales vel trivocales 
longa. Reliquae bivocales aliquando sunt breves, aliquando 
longae, interdum mediae ; adeóque nrmam non habent regulam, 
sed reguntur usu et authoritate. Quando dictio finit in 
consonam fortem erit mediae quantitatis ex natura sua ; 
tametsi ut dixi, poetae valeant necessitate ita compellente 
earn facere longam. Verum quando consonae hujusmodi 
fortes veniunt inter duas vocales ejusdem voculae sonant 
leviusculé et breviter ut coinne, tom^e, buitte, uiffve, CAif\f\e, 
mne, fin^e, etc. Verum si consona mollis subsequitur ad 
hujusmodi fortem in fine, syllaba tunc erit mediae quantitatis 
ut fpongc, etc. 

xii. Consona levis veniens ante consonam mollem, item 
mollis veniens ante levem anteriorem syllabam reddunt 
brevem nisi syllaba de se sit natura longa, ut ajac, 1*501 tc, 
m-ACfAiib. Consona autem dura ante levem posita praece- 
dentem syllabam imó post facit quantitatis mediae, ut tons, 
Za-ú^, etc., exceptis paucis ut ka^oa, bpAigoeb. Consona 
autem dura veniens ante levem anteriorem syllabam, nisi 
merit naturá longa, facit brevem, ut za^a, a^a, etc. 

De Prosodiá Hibernicd. 91 

xi. The diphthong eo is nearly always long except in a few 
words of which *oeoc, neoc are examples. The diphthong 
e-A is always either short or medium, except in the curotrtfpie* 
or verbal ending of the second person singular of the imperfect 
[and the conditional] as when I say x>& mbtiAitceÁ [or *oo 
ou-AilpeA] where it is long. But it is naturally short, unless 
it is found long among the other diphthongs and triphthongs. 
Other diphthongs are sometimes short, sometimes long, and 
occasionally medium ; so they have no fixed rule, but are 
regulated by use and authority. When a (monosyllabic) 
word ends in a strong consonant, it will be of medium quantity 
by its own nature ; still, as I have said, poets can make it 
long as the necessity arises. But when strong consonants 
of this kind come between two vowels in the same word, 
as in coinne, tomge, utntle, tnf^e, cdi^ie, wine, firige, etc., 
the syllables sound rather light and short. If, however, a 
soft consonant follows a strong one of this sort in the end 
of a word, the syllable will then be of medium length, as in 
f ponjc, etc. 

xii. A light consonant coming before a soft one, or a soft 
one coming before a light one, makes the previous syllable 
short, unless it is of itself naturally long, as in a\\c, r^oUc, 
imac\\ai?>. But a hard consonant placed before or even after 
a light one makes the preceding syllable of medium length, 
as ton^, Ua'ós, etc., except in a few words like dajvoa, bjvÁig-oett. 
A hard consonant, however, coming before a light one makes 
the syllable short, as in za^a, a^a — unless it should be 
long by nature. 

♦The term inr-jne found in ctu-o-i«fT;ne is variously spelt and is 
used in various senses — thus we find inpce, mnrce, mr-cm, nifcne, 
and all these written again (in later times) with 5 0uf5 e > etc.). 
It meant (1) speech in general, as fiAtin-A \r\xce=partes orationis, parts 
of speech ; (2) in particular a pronoun, e.g., Fr. Molloy's inline 
feAU)uii;te=possessive pronoun (as tno, -oo, etc.) ; (3) generally 
now for gender, being the grammatical term used to correspond to 
the natural term cinéAt=:sex : as piji-itifce (feAft-inr-ce ^masculine 
gender, bAin-infce=feminine gender. 

The root of inrce is probably innir-=tell, say, or perhaps an older 
form in--oirc containing a form 01 re connected with the Welsh dysg-u, 
to teach, Lat. disc-o, I learn ; Grk. dí-dáóx-aXog=a, teacher. The 
n found in last syllable (inr-cm, inr-cne) may be a trace of an old verbal 
ending (cf. innr-e imif in, cuisre ctnspn, or may only be a diminutive 

92 De Prosodiá Hibernicd. 

xiii. S praeposita cuicunque praecedentem syllabam voculae 
in qua est, nisi sit naturá longa, facit brevem ut toips, ceipc, 
etc. Praeterea syllaba brevis evadit media subsequente 
consona levi ante consonam asperam vel f ut tn.Ancoib, 
ninnpib. Item consona aspera praeposita levi anteriorem 
syllabam reddit brevem, ut eitne, bAitnib, etc. 

xiv. Syllaba quantitatis mediae nullam praecedit consonam 
simplicem seu unicam praeter solam m. Caeterum lectio 
authorum et usus te docebit quae Romanis procul positis 
non occurrunt. 

9. Addo alias adhuc regulas pro concordia, et sint sequentes : 

i. Deservit quaecumque vocalis ad concordiam intervenien- 
dam inter voculas modó initiative, et propriae vocularum 
literae principes sint vocales, ut alibi insinuaveram. 

ii. Si ad initiativam literam p sequatur n, tunc concordia 
non attenditur penes p sed penes quamcunque immediate 

iii. Sn non concordat nisi cum altera pb, neque p nisi 
cum f, concordant rg, fb, fm, fx>. Item rp, ft, pn. Con- 
cordat pn cum pn, necnon cum p. 

iv. Quando eclipsatur initialis voculae litera (excepta sola 
p) haud eclipsans sed eclipsata erit quae debet concordare — 
sic concordant be^n et tnbu.AC.AiU,, etc. 

v. Adverbium vulgó lAnmbéAptA non servit neque obest 
neque prodest ad concordiam ; etsi adjungatur dictioni 
nihilominus haud litera initialis ipsius adverbii, sed dictionis 
debet concordare ut alibi dictum de mo et T)o. 

De Prosodiá Hibernicá. 93 

xiii. S placed before any oth x consonant makes the 
preceding syllable short, unless it is naturally long, as toipj, 
ceipu, etc. But a short syllable becomes medium if a subse- 
quent light consonant precede a rough one or p, as m^ticoib 
rutin pm. So also a rough consonant coming before a light one 
makes the preceding syllable short, as eiupe, bAitpio, etc.* 

xiv. A syllable of medium quantity precedes no simple 
or single consonant except tn alone. But the reading of 
authors and custom will teach thee such things as do not 
occur to men living far off at Rome. 

9. I add a few other rules for alliteration, and let them 
be these : 

i. Any vowel serves for a concord — or alliteration — with 
any other vowel — only they must be initial and radical 
vowels of the words in which they occur, as I have implied 

ii. If an initial p is followed by n, then the alliteration 
is not determined by the p but by whatever letter immediately 

iii. Sh alliterates only with another pn, nor does p accord 
with f5 or ft» or ptn, or p*o — but only with another (pure) p. 
The same is true of pp, pt, pn. pn alliterates with another 
pn, and also with p. 

iv. When the initial of a word is eclipsed, then it is the 
eclipsed letter, not the eclipsing one, which is to alliterate, 
thus be-An and tnbu^CAitt alliterate. 

v. An adverb commonly called Mptnbe^ptA neither makes 
nor mars nor helps to make an alliteration, and if such be 
added to another word, nevertheless the alliterating letter 
will not be the initial of the adverb, but that of the other 
word which it qualifies, as was said before of mo and x>o. 

* It is curious that much as the author has said about the letter p 
and its powers, nowhere does he notice the distinction between the 
broad sound in pÁl and the slender in peol : fÁf rimes with bÁip 
and feAn alliterates with pviArt. Many circumstances seem to point 
out that the slender sound (sh in English) is not ancient but com- 
paratively modern — though, no doubt, the two sounds existed in Fr. 
ó Molloy's time. This sh sound does not appear to have existed 
in Latin or Greek, though modern Italian represents it by sci or see. 
The Russian has a separate character for this sh sound borrowed, 
I believe, from the Hebrew. French represents it by ch, German by 
sch, the Spaniards also seem to be developing it, for I have heard 
some provincials pronounce ciudadanos (citizens) — not with an s- 
or th- sound at the beginning but like the Irish syllable piú (—shoo). 
In words like tension, excursion where the English give the s the sound 
of sh, the French still give the original hard s sound. 


94 De Prosodiá Hibernicá. 

10. Rursús pro correspondentiis sit tibi prima regula : 
Prima : vocalis non correspondet perfecté nisi suae simili, 
etsi concordet quaecunque cum quacunque ut dictum est. 

Secunda : omnis consona concordat cum sua simili, imó 
correspondet, sed si in correspondentibus interveniant plures 
quam duae, non oportet quod correspondeant plus quam in 
duabus hinc inde, ut alibi dictum est ; ut enim correspon- 
deatur hinc voculae Ttjuiicc-psfieAtfi, sumcit intervenire aliam 
voculam correspondentem ei in medio per consonam c et p, 
sine eo quod interveniat vel c, vel 5, vel p ; tametsi inter- 
venientibus omnibus hinc inde, quod dimculter vel nunquam 
fieri potest, stetisset correspondentia. 

Tertia : nulli consonae licitum est non correspondere modo 
jam dicto, praeter interdum molles, de quibus supra meminisse 
oportet, sicut de dictis de potestate, eclipsi, et submersione, 
ne nimirum in correspondentiis sua vis non maneat cuique 
consonae juxta accentum. Bene igitur correspondent cAitceAp 
et -p.Aice.At) quia licet in prima vocula co-eunt duae consonae, 
nempé t etc in accentu tamen sonant unicam mollem. 

Quarto, : correspondentia requirit voculas convenire in 
numero syllabarum, quantitate, et ordine, non tamen ita 
et tanto cum rigore ut consonae non queant transponi, ut 
videre est in m-AcpAit) et fAtcAip. 

De Prosodiá Hibernicd. 95 

10. Again as to assonances, let this be the first rule : 
First : a vowel does not assonate perfectly except with 
its like, i.e., the same vowel in another word, although any 
one vowel may accord, i.e., alliterate with any other, as has 
been said. 

Secondly : Every consonant alliterates with its like [with 
the same initial consonant in another word], yea, even may 
rime with it, but if in riming words there occur more than 
two consonants, it is not necessary that more than two of 
these should correspond, as has been already said : so that 
to correspond with the word *0|\úiccf5|\eAtri it is only necessary 
that the other word should have a c and y in the middle 
without any c or 5 or -p, although if they could agree in all 
respects — which could rarely occur — the assonance would 
be all the better. 

Thirdly : Every consonant is allowed to rime in the manner 
already stated, except sometimes the soft ones (c, p, c), of 
which it is necessary to remember what was said above, 
as also about words, the powers of the letters, eclipses, elision — 
so that each consonant should have its own proper power 
in the rimes according to its natural pronunciation. There- 
fore cdicceAjA and -p^ice-AT) well assonate, because although 
two consonants t and c meet in the first word, still in pro- 
nunciation they sound as a single soft consonant. 

Fourthly : Assonance requires words to agree in the 
number, quantity and order of their syllables, not so much, 
however, nor with such rigour, that the consonants may 
not be transposed sometimes, as may be seen in tn^c^ró 
and fAiUAi^. 

96 De Prosodiá Hibernicá. 

Quinta : cum adverbio nunquam fit correspondentia neque 
una vox correspondere potest sibi ipsi nisi quando duplicem 
habet sensum, et sic dupliciter sumatur, vel quando inter- 
venit quaedam variatio vulgó b\\eAc&t>, ut cum vocula magis 
necessaria in semimetro, voculam quae vocatur uj\iAnn, vel 
aliquam voculam primi semimetri, vel alias voculas easdem 
sibi secum, seú ad se trahit, cum quadam elegantia, ut hic : 

-Af í Atl ÓAt.A01j\ ATI C-AtAOIfl, 

An óAtAoif 5-á corhtAtAoif. 
11. Haec sufficere possunt cuicunque mediocriter oculato, 
ut pcetarum sciat regulas, et Hibernicorum genera carminum, 
meo judicio omnium quae sub ccelo íiunt difíicillima. Veteres 
adhuc pro vitio habent nisi poéma claudatur seú finiatur 
eadem prorsus syllaba vel dictione vel phrasi qua incipit 
et vocant clatisuram. Verúm ego hoc non verto vitio, turn 
quia nullam video rationem, vel regulam sic prsescribentem, 
turn quia plerosque legi doctissimos et versatissimos qui 
hunc veterum non observarunt usum. 

De Prosodiá Hibernicd. 97 

Fifthly : With an adverb there is never an assonance made, 
nor does a word rime with itself, unless when it has a twofold 
meaning, and may be taken thus in a double sense ; or when 
some variation called b\\e&cA?> intervenes — as when a some- 
what necessary word in a semi-metre, especially that called 
the ufUMin, draws to itself another word of the first semi- 
metre, or other word like itself, as here : 

Ar í -An cac^01|\ An CAt&oM^,* 
&r\ óAtAoif 5>Á corn cacao 1|\. 

11. These directions will suffice for anyone moderately 
gifted with eyes to learn the rules of the poets and the varieties 
of Irish metre — in my judgment the most difficult to be 
found under heaven. The ancients also consider it a fault 
if a poem does not close with the same word or syllable or 
phrase with which it began, and this they call the closing 
[•oúnA-ó or ce^ngAt ?]. But I do not consider this a fault, 
not only because I see no reason for prescribing such a rule, 
but also because I have read very many learned and accom- 
plished poets who have not observed this practice of the 

♦This word cAcAoifi is a good one for a quibbling or punning verse, 
on account of its many meanings ; (1) a chair or seat — from Lat. 
cathedra, W. cader ; (2) a city — another but rarer form of the pure 
Irish word cAcAift, W. caer ; (3) the proper name of a man, CAtAoift 
("Cahir") for the earlier Cat:ai-ji or CAteft for cAt-fe-fi=man of war, 
warrior, battler. It may also be a compound of aoiji, a satire — 
cAtAoiji would then=a battle -satire ; the last word of the above 
couplet may contain the key to the puzzle : it is a compound of coiii- 
i&i&oiy, of which the last part itself=-oo-At-Aoi|i, a reproach, reproof. 

98 De Prosodiá Hibernicá. 

CAP. XII. [xxv.]. 

De vulgari versu et compositions 

1. Vulgarem voco versum quo vulgus utitur, Hibernis 
Atif An, seíi bujvoún seú ce-dtf axx\a ; item c-Aoine seú ctn^ e^t). 

2. Au|Un autem quatuor constat quartis inter se quantum 
ad Amur et auricularem sonum, omninó similibus, ita ut 
nullum excedat alterum, et in nullo sit repetitio ejusdem 
voculae, ut taceam articulos, nisi forte elegantiae gratia. 
Quartum itaque ipsum quodcunque quinque circiter constat 
pedibus ; primus erit unius vel duarum syllabarum, in versu 
saltern simplici, et non duplicato ; quae non correspondeant 
nisi primis pedibus subsequentium quartorum. Secundus 
pes et tertius invicem correspondebunt ad aurem, et singular 
duabus constabunt syllabis ad minus. Quartus duas ad 
minus continebit syllabas nee correspondebit nisi quartis 
pedibus trium subsequentium quartorum. Quintus pes 
ordinarié erit unius syllabae. 

De Prosodiá Hibernicá. 99 


Of Popular Poetry and of Composition. 

1. By popular poetry I mean the verse used by the common 
people — chiefly of two kinds — known to the Irish as (1) the 
&X)\\Ár\* or song, called also bujvoún or ceAtpAmA, and (2) the 
c-Aome or elegy — called also zw^eAX). 

2. Now the (stanza of the) Ab^Án or song consists of four 
lines alike in every respect as to Atntif and other rimes, so 
that no line exceeds the other, and in none is there a repetition 
of the same word — I say nothing of articles and other un- 
emphatic words — unless, perhaps, for the sake of elegance. 
And each line in itself consists of about five feet,f the first 
of which will be of one or two syllables — in a simple verse, 
at least, not in the double — and this foot will rime only with 
the first feet of the subsequent lines. The second and third 
feet will rime with each other, and will consist of two syllables 
at least. The fourth foot will contain at least two syllables, 
and will not rime except with the fourth feet of the three 
subsequent lines. The fifth foot will usually be of one 

* The older and more correct form is ArhftÁn, which may be from 
ATTTfiA (or Ath-fiAt) which sometimes means a panegyric, sometimes an 
elegy. In form it is a diminutive, like the Latin cant-ic-um, cantio 
(whence French chanson), cantilena, etc., buji-oún is a late Norman word 
from Fr. bourdon, whence also Eng. burthen, burdon : it is still used 
in Munster for a song. CeAtfiArh, ceArfiAtriA (or ceAtjiAiriie) is used in 
three different senses : (1) the fourth part, i.e., a line or quartan of a 
four-lined stanza : this is the very oldest sense (as applied to poetry) 
and it is used by our author in this sense : see Chaps. I. and II. ; (2) 
the fourth part of a four-stanza song, hence any stanza or "verse" 
of a song ; (3) a short song, any song, as used here by Fr. ó Molloy. 
This last sense is rare. CuifteA-o is more correctly spelt ctnjieArh 
(see SeAcc SÁfi-óÁncA 5 Ae "° 1 t5e). 

f Notice this is the first time that the author speaks of feet (" pedes ") 
in treating of Irish poetry — in all the previous chapters he has spoken 
of syllables only as constituting the cóniAf or metre. Observe, too, 
that the author speaks unreservedly of feet of one syllable. However 
shy modern scholars may be in admitting such in classical Greek or 
Latin verse there is no doubt whatever that we have monosyllabic 
feet in modern Irish verse. 

100 De Prosodiá Hibernicá. 

3. Dixi ad minus, quia hujusmodi versus alii sunt simplices 
de quibus jam sum locutus ; alii veró vocantur Hibernis 
•oúbAixA seú duplicati, quorum pedes plures ad libitum 
authoris admittunt syllabas. Et sane suas habent diffi- 
cultates, nee tamen reputantur nisi elegantiam habeant et 
gratiam vel conceptum, quandoque autem sunt prsegnan- 
tissimi sensus, ut in sequenti videre est, a quodam facto in 
separatione sui olim Achatis : 

A oLAitUofv tÁjAó t>á ^cpémteAf no, 

t)í 1T> LÁirifuot 50 t)ÁnoA '-pD^f fjeut-A tio, 

5101!) "OALLftro luce Áittiteif mé 5-án pof, 

50 |\ti5 pÁcc|vui5 mo t>& t)eAj\c 50 ti-éij\e tif ! 

Quam autem prsegnantis sit sensus hujusmodi plebei versus 
rite compositi, conjice ex sequenti Latino fermé aliud con- 
tinente : 

Lingua nocet multis, nam coecutisse Maronem 

Nasonemque loqui fecerat, et reliqua : 
Erravere tamen qui me finxere strabonem, 
Non qui nunc coecum, luce abeunte 

De Prosodiá Hibernicá. 101 

3. I have said at least — (with regard to the two syllables 
of the second and third feet) — because some verses of this 
kind are simple, of which I have just spoken, but there are 
others called x>úb&lzA or double by the Irish, where the feet 
admit of more syllables at the will of the poet. And truly 
they have their own difficulties, but they are not in much 
esteem unless they have elegance and grace and thought — 
unless they are pregnant with sense, such as may be seen 
in the following specimen composed by a certain writer on 
separating from his former Achates : 

A DtÁittici|v LÁjAó x>Á ^cferate^ fit), etc.* 

Literally : 
O fair and loving letter, speed thou on to Banbha ! 
Tell them my sad message — if they can believe thee — 
Tho' my foes have blinded me beyond all knowing, 
Only now at last have my two eyes gone to Erin, 
And I no longer see ! 

How pregnant the sense of such popular verse may be 
when duly composed you may guess from the following 
Latin version made almost literally from the Irish : 

Lingua nocet multis, nam coecutisse Maronem, etc. 

Literally : 
The tongue injures many, for speaking made Maro blind, 
Ovid, too, suffered for speech, and I could name many 

Still they have erred who declared me oblique-eyed, 
— Not those who say I am blind, now that my light is 

all gone ! 

* This example of a song metre by the author himself is not the 
measure described by him above — it is one of four feet (not five), the 
last foot being monosyllabic, the first three trisyllabic and of an " amphi- 
brach " nature. Keating, O'Rahilly, MacDonnell and others wrote 
songs in this style and metre. Moore imitated the metre in " Let 
Erin remember the days of old " — at least in the odd lines. It is 
curious that the author in turning his song into Latin does not use a 
lyric measure, as one might expect, but the elegiac. He was thinking, 
perhaps, of its mournful nature and subject rather than of its form. 
The whole of the Latin poem is given at p. 111. 

There are but two types of Irish song noticed here — or, rather, there 
is one type described and another illustrated. But modern Irish 
songs exhibit great variety of style and metre, and a thorough study 
of these metres and styles would be most interesting and valuable 
to all lovers of our native music, poetry, and literature generally. 

102 De Prosodiá Hibernicá. 

4. Dixi ad minus, aliquando enim pedes primi quartorum 
constant tribus syllabis, et in duplicatis secundi et tertii pedes 
constare solent tribus ad minus syllabis si non pluribus. 
Praetereá ad simplices et duplicatos versus nonnulli, vel 
elegantiae gratia vel alio motivo adjungere solent alios duos 
pedes sequentes uniformiter ad quodlibet quartorum, ita 
tamen ut hi duo pedes, alter bissyllabus alter monosyllabus, 
vel plurium syllabarum, neque inter se correspondeant, neque 
cum aliquo pede istius quarti ad quod sequuntur, bene autem 
cum consimilibus pedibus ad reliqua quarta subsequentibus, 
vel etiam, ut unusquisque pes sit bissyllabus, et hoc vocant 
-plAc -AnTn-Aij rhAtUvij; ; ut si ad primum quartum praefati 
versus adderetur At>uf nó t&ll, ad secundum veró aj\ -puo 
tiA ngteAnn, ad tertium rnAf gut no peati, ad ultimum 50 
bun Af rh e-Ait, quod et fieri potest in quibuscunque versibus 
etiam duplicatis modó sic voluerit author. 

5. Alterum vulgó cAome solet fieri tanquam naenia in 
alicujus obitum, tametsi fieri posset alia in materia, in 
scilicet vel laudem alicuj us vel vituperium ; et quilibet 
versiculus ipsius constare solet quatuor tantum pedibus, 
quorum quilibet ordinarié erit bissyllabus, neque requirit 
primus pes habere correspondentiam cum ullo, neque secundus 
vel tertius, nisi quod inter se quasi saltern correspondentiam 
quoad sonum et aurem servent ; postremus autem pes debet 
hoc modo correspondere cum postremis pedibus omnium 
subsequentium versuum, ut : 

ftUAfACAC ^ACtflAf f ACCrhA|\ éACZAC 

De Prosodid Hibernicd. 103 

4. I have said of two syllables at least, for sometimes the 
first feet of the lines consist of three syllables, and in double 
verses the second and third feet usually consist of three 
syllables at least, if not more. Further, both to simple and 
double verses, some for greater elegance or for some other 
reason, are accustomed to add two more feet of uniform 
length to each line, so that these two feet — the one being 
a dissyllable, the other of one or more syllables — neither 
rime with each other, nor with any other foot of the line 
to which they are added, but make a good assonance with 
similar feet in the remaining lines — so that each foot (of 
these additional feet) may be even a dissyllable. This 
addition they call pLac ax\t>\a\% rtiAttAij ("a yard to the old 
lay " ?) : as would be done if to the first line of the above 
Irish verses were added Abur nó tail, to the second a]\ -ptro 
tiA n^teAnn, to the third rruÁr 511c nó yeAll, and to the last 
50 bun A]\ me All ; which may be done also in most verses, 
even in those called double, if the author wishes. 

5. The other kind (of popular verse) known as the cAome 
is usually composed as a naenia or dirge for the death of 
some person, although it may be made of another character, 
namely, for a man's praise or dispraise. Every line of a 
cAome usually consists of only four feet, each of which will 
ordinarily be a dissyllable, nor does the first foot require 
to rime with any other, nor the second and third except with 
each other as to sound and syllables ; and the last foot should 
rime* with the last feet of all the subsequent verses, as : 

RUAf\ CA6 fiAtmAp pACZlftA\\ éAtZAt, 

O'Ruarc the noble, mighty, glorious, 
O'er troops of foes in fight victorious ! 

* It is to be noted that though the author in this chapter on Popular 
Poetry says much of correspondentia (assonance or rime) he says not 
one word of concordia, i.e., «Aim or alliteration. The fact is, the 
more assonance was cultivated the less the initial harmony or alliteration 
was regarded ; and whilst this -uAim or alliteration was an essential 
in a line of Irish verse in -oÁn -oi^ieAc, in the folk verse on the other hand 
it was only an ornament. Some of the popular poets used it very 
sparingly, others used it largely. But from the earliest times «Aim 
or initial harmony was a characteristic of Irish style both in prose and 
verse : one of Fr. ó Molloy's own examples of the vulgaris versus — 
UuA-pcAc tiAttriAfi, etc. — shows almost as much observance of allitera- 
tion as of assonance. 

104 De Prosodiá Hibemicá. 

Aliquando tamen pedes intermedii transeunt in trissyllabos, 

tTlóince^ó, tn^óAipe-Aó,, e15ne.dc : 
Sic de reliquis. 

6. Lascivia, impuritas, dirae, obtrectatio, maledicentia, 
mendacia, discordiarum et contentionum sementatio vitia 
sunt non artis poetical, sed perversissimi ejusdem usus, 
dixerim abusus. Contra quos divinus ille vSancti Hieronymi 
ex specu Bethlehemitico auditus est rugitus (in quintum 
Amos) : Non debemus sequi fabulas poetarum, ridicula et 
portentosa mendacia, quibus etiam Coelum infamare conaníur, 
et mercedem stupri inter sidera collocare : dignos utique quos 
Minerva casside sua et cuspide ex toto arceat Parnasso. 

7. Placent pise qusedam Hibernorum compositiones ut 
famuli Brigida^ in mundi contemptum, Donchadii Dalaei 
cognomento Magni in laudes Dei, et Deiparae Virginis semper 
Immaculatse ; aliorumque devoti Hibernorum rhythmi, quales 
apud Sanctos Doctores, et Bernardum de Passione Christi, 
et Seraphicum de Philomela et similia : inter qua^ locum 
habet sequens compositio íideliter translata instar Dialogi 
ad honest am juventutis diversionem inter Tityrum et N. 

De Prosodiá Hibernicd. 105 

Sometimes, however, the intermediate feet pass into tri- 
syllables, as in — 

ITIomceAC, tnACAijAe.dC, At3Uinne.dC, éigneAC — 

and so of the rest.* 

6. Licentiousness, impurity, murders, vile calumnies, scur- 
rility, lying, the fomenting of quarrels and discords, are faults 
not of the Art of Poetry, but of its perverted use — let me 
say, abuse. Against those who thus abuse it, this loud 
condemnation was once heard from Saint Jerome speaking 
from his cave at Bethlehem (on the fifth chapter of Amos) : 
We should not be led away by the fables of poets, or by their 
ridiculous and monstrous lies in which they do not hesitate 
to defame heaven itself, glorifying and raising to the stars even 
the grossest depravity. Such poets as these deserve, so to 
speak, to be cut off entirely from the sacred mount Parnassus 
by Minerva armed with spear and shield. 

7. Many pious compositions of Irish poets give pleasure, 
such as that of Giolla Brighde (0 h-Eoghusa ?) On Contempt 
of the World, those of Donnchadh ó Dálaigh, surnamed the 
Great, on the Praises of God, and On the ever Immaculate 
Virgin Mother of God ; besides devout poems of other Irish- 
men like those found in the writings of the Holy Doctors — 
as that of St. Bernard on the Passion of Christ, that of the 
Seraphic Doctor [St. Bonaventure] who wrote the poem 
called ~Philomena\ — and others like them. Amongst these 
the following composition finds place, written dialogue- wise, 
for the innocent amusement of youth, and here faithfully 
translated (from the Irish) : 

* The treatment here given to elegiac poetry is very brief ; but though 
there are many elegies written in the metre above described, for which 
reason it may be called an elegiac metre, there is hardly one that may 
be called the elegiac metre of Irish verse — for elegies, ancient and 
modern, are found written in all kinds of different metres — as a reference 
to any collection of Irish poetry will show. 

f Fr. ó Molloy uses the older form Philomela (the nightingale), but 
the actual word used by St. Bonaventure is Philomena, a medieval 
corruption of the classical word. The author used it also in a figurative 
sense, meaning a devout soul. The poem occurs in the sixth vol. of 
the Saint's Opera Omnia, Moguntiae MDCVIII. [This reference has 
been kindly cleared up for me by a well-known and learned Irish 
Franciscan, Fr. A. D„ O.F.M.] 

106 De Prosodiá Hibemicá. 

Tityrus : Colloquamur ambo 
Ambo ambulemus, 
Amplexemur ambo, 
Ambo osculemur. 
Amputate vites 
Minus gracilescant, 
Ne labascant vires 
Laborantes cessant. 

N. Abi tu si velis 
Mihi est quod 
Abi, poenitebit, 
Imitare sagas ! 
Ego te defiebo, 
Stultum atque reum — 
Semper hie studebo 
Et orabo Deum ! 

Et haec de his pro nunc sufnciant levidensibus. Cseterúm, 
si in aliquo hic forsan defeci, vel excessi, qua^so excuser, a 
quadraginta et amplius annis inter exteros procul positus a 
Patria, Patriis monumentis, et magistris qui alias mihi omnem 
possent tergere caliginem. 

Sit nomen Domini benedictum ! 

De Prosodid Hibernicd. 107 

Tityrus : Let us go forth walking 
Talking as we wander, 
And embracing kindly. 
Seek the wild wood yonder ! 
Prune thy vine's stray branches 
And 'twill grow less meagre, 
To rest at times from labour 
Toilers must be eager ! 

N. Go thy way to pleasure, 
I must go to labour, 
Go — thou wilt repent it, 
Play like idle neighbour ! 
I shall ever weep thee 
Voice in blame upraising, 
I still love to study, 
Ever God's works praising i* 

[At the end of the author's Soyun-o 50 tiAof 05 ^tif eAZA 
Oitém x\a TiAom — given here at page 112 — he finishes the whole 
work, Grammar, Prosody and Poems, in a few words, of which 
the following is a translation] : 

And now what I have said here of these trifles must suffice 
for the present. But if in anything I have here failed, or 
gone too far, I pray I may be excused — having lived forty 
years and more amongst strangers far removed from native 
land, from native books and records, and from masters who 
could otherwise help me to clear away every obscurity. 

Blessed be the name of the Lord! 

* The Latin poem — the first and last two quatrains of which are 
quoted above — covers thirty-two pages in the first edition of the 
Gvammatica, and extends to over 700 lines. It is too long to be 
given here, and I must refer readers, therefore, to the original work. 
Perhaps the original Irish poem, of which it is a translation, may still 
exist, and be yet brought to light. It was probably a medieval com- 
position, perhaps written in -jnonnAji-o — for the Latin seems an imitation 
of that metre. The poem has many of the qualities of Irish verse — 
especially its alliteration and rime-system. 

108 De Prosodiá Hibernicá. 


[These poems are mostly Fr. ó Molloy's own compositions, 
and occur in the course of his Prosody in illustration of Irish 
metres. I thought it best to collect them here, together 
with the Latin versions given by the author to some of them]. 

Cum ATI AtAít pAT)ttA1C umiAt,* 
t)fvÁtAif\ bocc *o'ó|Vo Say\ Pf\omriAr : 


['Oeibi'óe é reo]. 

^tAine tió các tú mtj|\ tjuAt 
tAige a ccfÁcc nó Uif\iAt — 
SubÁitce 5AT1 céitn a\\ cút 
Ufvén j\e mibÁilce a •oeAóLú ! 

thJAn "oo tionfsn-árh a cclÁf\ Cumn 
A pÁm lontfmm rmc Gtmitnnn, 
LeAC An fiteAt) meAf Af rnó 
ílí -pe^f "00*0 cineA*ó cLaocIó. 

T)o fioffAt) mé a ftomne fiAn 
OonnAf TnotCAn le Uaíía — 
An btÁitbé rheAn 5 An rheinse — 
t)eAn An 5f\Airnéin ^ 01 * 0611 ^ : 

A5 nonbeAnsA-o T)on 5615 51 1 
"Otró téAn ni'-pneAgnAt) ón in§in, 
'Stío but) rítbmn te m'ó a fior, 
Aoibmn An ^lon -o'Án géittiof ! 

1TlAit ot rí An po^Amrn -piAt 
SojAintn nAó cfAotAfv UiniAl — 
t)u CAnAit) 5 An bérni 50 bÁr 
^au rhAlAific é a n-AonóÁr ! 

* Father Patrick Tyrell — as already stated — was Ex-diffinitor General 
of the Franciscan Order, and one of the Censors to whom Fr. ó Molloy's 
work was submitted before publication in 1677. 

De Prosodid Hibemicd. 109 

Latiné : 
Nee te candidior Tyrel, nee firmior extat, 

Cui comes ut virtus, nescia fama mori. 
Ccepisti, stabis, nee declinaberis hilo 

Cor quia Patricii sors variare nequit. 
Proposui nuper num possit fore Thalia 

Grammatica agnomen flectere lege tuum. 
Cunctatur, dubitat, pallet, rubet, haesitat, alget, 

Hsec hilari tandem voce Thalia refert : 
Esse Tyrel dicas indeclinabile nomen, 

Mutari nescit, casibus ergo caret ! 


T)on pUiincéAt)AC iarU\ pine SAtt.* 

[SéAvriA é feo]. 

A £>tAit x\a tntiAt) a ptúif\ pttmcécc 
A lAftAi-ó 5tófrhAi|\ priit) J^&lt — 

'ÓAOltt Af\ CUIJU'O DfUAJtilAjA t)|\15*Oe 

tTHle bUAjAn \ ctntte tA\X ! 

T)uic f-Ar» Hóirh pÁitxeA'ó if piceA*o 
Aif'oeA'ó t)Ár\A UiAi'óceAfv tmn — 
A tÁm aw ceiju; tiAC -p-Ann péiti 
A tArm ati eic jtégit £irm ! 

T)Á xtru^A'ó T)ia 'óúinn triA^ gt^f^ 

PA1Cf1t1 c'AJAlt) A CCtÁfA Cuij\C 

ílíf\ fteA^ linn acc fin rnuf f aojaL 
T)Arh 'f" 00 ™ "ófeAtn níj\ £)ao§aL bjuiro. 

SÚT) 0|\t11t) Of ÍOÓTlAf m'oCOA 

SLÁwce cittín ctéib tno ctntn — 
1TlA|\f ha mot) fAojAtCA fíte, 
AoítoaIxa fgot cj\íce Cuinn ! 

Latiné : 
Flos procerum, Fingalle Comes, Plunchette planeta, 

Mille tibi post hoc Brigida festa paret ! 
Munificae justaeque manus, cantaris in urbe 

Fraenator, cygno candidioris, equi ! 
Te si forte domi videam, nihil ampliÍLS optem 

Nulla mihi restet poena, nec ulla meis. 
Deprecor ex toto binam tibi corde salutem 

Ubera cui Pallas Jupiter arma dedit ! 

* See p. 51 ante. 

110 De Prosodiá Hibernicd. 


Lingua nocet multis, nam coecutisse Maronem 

Nasonemque loqui fecerat, et reliqua. 
Erravere tamen qui me finxere strabonem, 

Non qui nunc coecum, luce abeunte mea. 
Dum simul extitimus, simul hanc dum suximus auram, 

Te, caput hoc, oculos credidit esse suos. 
A me te tandem disjungit Hibernia mater 

Sic oculis sobolem, priva Ierna suam. 
Mitius o mater, seniori irascere proli, 

Et me, non sine me, Mater abire sinas. 
Abnuis. Hie potero tamen absens dicere : Ierna est 

Patritiis oculis tota videnda meis. 
Ast ego quid faciam ? patiar mihi nunc et in urbe, 

Quamvis non liceat flere, dolere licet. 
Et doleo et nequeo fugientem flere Tyrellum, 

Causa subest — oculos abstulit ipse mihi ! 

AW tXáS.f 

IDormcAO mófv ó T)ÁLdi$ po tA\\ : 
[ÓgtÁóAf fVArmtngeAóCA tnói|\e]. 

Ufitifv &cá 45 bfidic aja mo bÁf 
510^ acAit) "oo gtiÁt im x>\m, 
Ct^AS 5 An A cc|aoóa f\e cfAtin — 
Ax\ x>\aX)A\, av\ cl-Anti 'fA crmrh. 

/Au co|\p Ati z-AriAm ay\ fpj\é 
A^ nmit "OArh 1 cc^é mufv óÁó, 

OfXJVAt) AZÁ b|\A1C AW Ufl1.A1J\ 

'S-Af T>eirhin 50 tnbi-di* 50 bjvát. 

tlí tiob^At) AomneAó *oon Cfutif\ 
T)ori "Oíf oite 510*6 mt ctaon, 

* This is but a translation of an Irish original which the author, 
however, does not give, but from what is said at p. 97, the Irish appears 
to have been an ArhjiAn. 

f This is the only piece not by the author, and I give it here partly 
because it is one of the author's examples of ó^lÁÓAf, partly because 
it is somewhat rare — the version given by O'Daly in his edition of 
UA-Ó5 ^Ae-oeAl-Ac being imperfect. It is sometimes attributed to a 
later poet, giotlA-bftij-oe ó h-eojupA (early 17th century). 

De Prosodiá Hibernicd. Ill 

An cint) T)o noitpeA-ó 'nA 5615 
T)nóib An a ccuro pém aj\ Aon. 

An T)1AbAt Af T>0}YÓA T)Alt 

An peAn teif nAc Ait acc otc, 

An An AriAm roitbin réim 

Tlí jeAtDA ré An rpneit) fA conp. 

TIa cnurhA 510*0 Arh5An rut) 
'5-Á scuntAn mo cut 1 5cné 
T)ob 'ipeA|\|A tni aca mo conp 
11Á m'AnAm bocc if mo fpné ! 

T)o bfeAnn te mo ctomn mo fpnéi'ó 
*Oo belt aca pém Anoct) 
*ÓAmfA 510*0 pA5tif a n5Aot 
Ho m'AnAm AnAon 'f™ conp. 

A Cniofc "oo cnocA*ó te cnAnn 
'S*oo 5onAii) te t)Att 5An mt — 

Ó CÁ1T) A5 bf\Alt Af\ mo f tAT) 

1r cnuAj; 5An 5AT) A|\ An cniun ! 

sotttUT) ó 'OitneAtDAó ntiAtnA 50 clan coinn : 

[This was written by Fr. ó Molloy in 1676, and is given 
at the end of his tócnAnn nA 5Cneit)meAc, published that year 
in Rome.] 

(Deibi'óe é peo). 

A teAbnAW nío^A ón ítónri 
AcÁ tib téigeAnn tÁncóin, 
5o n-1nif bpó'ogtAm nA bpionn, 
tjur mitir *o'ó5Aib éifuonn ! 

1ut cneiT)im CAtpnum cnoi'óe 

T)í ogttum gtAn cntiAfAó Cníor'ooi'oe 
TViAnr-mA ftnAgAi'ó jntbAn gwnc 
Ap cttiAnAib ujj-oAn n-ofvouinc. 

<CeA5Af5 ÓníopT) nA puA-o 'pnA peAn 
CiAtt nA bponc *oÁn coin cneiT>eAm, 
THot) nA 5CtiAn bpAttfA *oo ótAoi 
'S'oot te AnnfAót) T)Án n-Áijvocín. 

A5 fnA'ómA cAiTineAm pAn 5cníc 
A5 coinmeAfc tntc if pAinbnig 
A5 cniAtt CA5uit) 5e be beAp 
ílí n-AbAi*ó a iac 5An c'omeAp ! 

112 De ProsodidfHibemicá. 

1Tleic eAgUnfe 51 1AT) Ann 

UAnbAó "oon tucc nof teij;eAnn 
An ojvo s-Aifgit) 51 be beAf 
t)An bfAicfin ní né AirhteAf. 

te ctAmn ttlíte t>Ar\X)A bÁm 
An ccúf tAbAin a teAbnÁm, 
11a bí coitóeAf, coitt a ccoiL 
OinceAf nomn m niogjwo. 

An neAc r>o f tiocc Jaoi-oiL $tAif 

Y\Á Celt 5A6 1tlt T>A bfUAfAlf 

fló An feAn^AttAib fóit) nA brionn, 
tén óeAngtATíAn 015 6inionn. 

UAbAin fóf "o'eicrib urhtA 
T)on AOf eAgnAix) eAlu'onA, 
ConnAitbe Af t)ú Leif m n^neim 
Unú a^ a ccornAifse cinnemi. 

tTleic oltAtfmm 1nri pÁiL 

'bé'm noriitnb lÁn T)o LutjÁin, 
'Oot) coirhneic ó beot 50 bét 
X)u*ó oifvóeinc c'eot fAn oitén ! 

TIa cinneAi!) fib ruAf *oo neAó 
teAtunom nÁ b-iAnn a\k émneAó, 
An fonn a^ ^ac cAob ciomcuit 
tlonn te 5AC Aon T)'6inionnctiib ! 

An n^-Aoi'óit a\\ b£ionngoilt fém 
"beAnnocc cné iMojnAif "oóibfém 
t)ein mo beAnnACc 50 bfém b£Áit 
"beAnnocc X)é tib a LeAbnÁm ! 


sorttnt) 50 n-Aos 05 A5us eAUA oitéin nA nAorh : 

[These lines are given at the end of the Grammatica Latino- 
Hibernica, and are probably the last Irish verses composed 
by the author.] 

('Oeibi'óe é reo). 

UnuA$ >oAoine a\^ *oit bicni, 
lAn cctAoi'óe 'o'-ptiAt Aibgicni, 
An AmóeAf Af tón t>o tÁn 
Ambf eAf Af rnón An rhónÁn ! 

De Prosodid Hibernicá. 113 

"P(yotA tm "do cuaitd rriAn ccioc 
An A%A\t> rtuAij a rmrion, 
Hín An oippveAi) An £umn 
poincraeAt gtAn nó po£Un rn. 

TIÍ ting ^ 01 * 11 - H^ 01 *^^ P é1n 
ílí tAbnAiT» í s^n Aombéim, 
Hi téigra te cÁ^Ató coin 
Cnéi^it) ní pÁgATO onóin. 

t)e^5 a in At) An riA teAbnAib, 
An 5tón ceAnc a cceiteAbnAib, 
tlí bneAt "óAon "o'piAntnb "Pail 
An niAjtut ctAon *oo congbÁil. 

teujAt) nAinn t>a n-iAnn onnA 
'Ha cceAn^An!) rhín rhAtAjvoA, 
*OioríiAoin oc An eAfbAt) iat> 
T)eAf5A > ó nA S50C ón Scicia. 

£itt AnofA a Aor m'AnmA 
Há bi 50 "oiAn > oo$AnmtA, 
'SnAó ciAn ó 6AtfAointiof Cuwn 
50 mbiA An c-AtAoibneAf Aguinn. 

t)iAi*ó nA bningne 'nA tnbotA 
Ha bpíoncA biAit) bÁnfnotA, 
Cntnc ú|\a nA njeAlponn nglAn 
A bpeAnonn T)únA TDeAtgAn. 

"biAt) An jAoi'óeAtj -pA rheAf rhón 
An At ctiAt nA bpteAfg bpíonót, 
"but) mAc a piAn but) ^Aot ^nmn, 
'Sjac íaoó A5 cniAtt pÁ ctiAinim. 

T)eAn?;pA cneAbAó a tutcA 
OnpA sniAn a ^lAfóunlA 
Hí "oo óAn -o'Oinmn An c-eA*ó 
SoóAn nAó éiT)in t/AineAtn. 

CtnnpeAt) coniAom An teit cCtnnn 
'SAn lAnrrnAt) Oittit Óluirn, 
t)onb A5 cnémcneAf^nAt) An ctnn 
-Af ton^ ^pémeAfA "pAnrui-ó. 

Aifce 'óenncA "ouAn ir x>Án 
^eib teAn 50 tén im LeAbnÁn, 
Of cobnAib TluAmA nA nAt, 
ObAin fCUAmA An fAotAn. 

114 De Prosodid Hibernicd. 

Sm Ó145AT) ón mbfiÁtAi^ mbocc 
^áyy)azz be^5 bu-dit) if be-Annocc, 
te fUAgtnt fgot fA fArhint 
TM.Arh.4itt tiA mboc rnbttoTMtfiuit. 

Son n^ ngot.Ai'oe 50 gttmn 
CurhACC con ran iac pen!>Lim 
CturhmtijAt) bÁtAt) ^An bnui-o, 

T>o rhe-At nA rAoite te fe^t, 
Aor 05 ir e-AC-d 5^01*0 eAL, 
T)o tdíoía a tóúttAnn 5^n Iua?> 
pónl^nn ^n ttíogttA ttó tfu-dj !